1,065 185 15MB
Pages 278 Page size 472.5 x 675 pts Year 2008
Second Edition ABRAHAM H. MASLOW VAN NOSTRAND REINHOLD COMPANY NEW YORK CINCINNATI TORONTO LONDON MELBOURNE This bo
8,886 1,666 956KB Read more
Feminist economists have demonstrated that interrogating hierachies based on gender, ethnicity, class, and nation, res
1,252 311 2MB Read more
Technology, E-learning and Distance Education Second edition Technology, E-Learning and Distance Education is the essen
2,309 1,065 874KB Read more
Edited by Wai-Kai Chen University of Illinois Chicago, U.S.A. Boca Raton London New York A CRC title, part of the T
964 153 10MB Read more
Edited by Wai-Kai Chen University of Illinois Chicago, U.S.A. Boca Raton London New York A CRC title, part of the T
1,112 315 7MB Read more
IdlVgY 6EhnX]dad\nd[ JcXZgiV^cin Psychoanalytic Inquiry Book Series Volume 27 ER4786X.indb 1 10/19/07 6:42:56 AM
1,121 513 2MB Read more
Toward a Science of
Toward a Science of
DISTRIBUTED LEARNING E D I T E D
Stephen M. Fiore Eduardo Salas
Copyright © 2007 by the American Psychological Association. All rights reserved. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, including, but not limited to, the process of scanning and digitization, or stored in a database or retrieval system, without the prior written permission of the publisher. Published by American Psychological Association 750 First Street, NE Washington, DC 20002 www.apa.org To order APA Order Department P.O. Box 92984 Washington, DC 20090-2984 Tel: (800) 374-2721 Direct: (202) 336-5510 Fax: (202) 336-5502 TDD/TTY: (202) 336-6123 Online: www.apa.org/books/ E-mail: [email protected]
In the U.K., Europe, Africa, and the Middle East, copies may be ordered from American Psychological Association 3 Henrietta Street Covent Garden, London WC2E 8LU England
Typeset in Goudy by World Composition Services, Inc., Sterling, VA Printer: Data Reproductions, Auburn Hills, MI Cover Designer: Berg Design, Albany, NY Technical/Production Editor: Kathryn Funk The opinions and statements published are the responsibility of the authors, and such opinions and statements do not necessarily represent the policies of the American Psychological Association. Library of Congress Cataloging-in-Publication Data Toward a science of distributed learning / edited by Stephen M. Fiore and Eduardo Salas.—1st ed. p. cm. Includes bibliographical references and index. ISBN-13: 978-1-59147-800-3 ISBN-10: 1-59147-800-6 1. Distance education—Computer-assisted instruction. 2. Computer-assisted instruction—Social aspects. I. Fiore, Stephen M. II. Salas, Eduardo. LC5803.C65T68 2007 371.3'58—dc22 British Library Cataloguing-in-Publication Data A CIP record is available from the British Library. Printed in the United States of America First Edition
o> ^ 4)
£« Si Q
o M-i £ O
1 PQ I
'a "O " • 5 G g 51 gg ^ CN '»
« I 8
1 £ T-H
js 2*-i -JO
To Paul and Mary Fiore, whose passion for knowledge served as an inspiration to their family.
Theoretical Models Emanating From Distributed Learning in Organizations
Moving Us Toward a Science of Distributed Learning Eduardo Solas and Stephen M. Fiore
A Theory-Based Approach for Designing Distributed Learning Systems Steve W. ). Kozlowski and Bradford S. Bell
Applying a Social Capital Perspective to the Evaluation of Distance Training Kenneth G. Brown and Mark E. Van Buren
A Meta-Analytic Investigation of Learner Control: Old Findings and New Directions Kurt Kraiger and Eddie Jerden
Distributed Teams and Distributed Team Training
Advanced Distributed Learning for Team Training in Command and Control Applications Barry P. Goettl, Alan R. S. Ashworth III, and Scott R. Chaiken
Distributed Mission Environments: Effects of Geographic Distribution on Team Cognition, Process, and Performance Nancy ]. Cooke, Jamie C. Gorman, Harry Pedersen, and Brian Bell
Cognitive Processes and Products in Distributed Learning Environments
Narrative Theory and Distributed Training: Using the Narrative Form for Debriefing Distributed Simulation-Based Exercises Stephen M. Fiore, Joan Johnston, and Rudy McDaniel
Five Features of Effective Multimedia Messages: An Evidence-Based Approach Richard E. Mayer
Engaging and Supporting Problem Solving Online David H. Jonassen
Question-Asking in Advanced Distributed Learning Environments Robert A. Wisher and Arthur C. Graesser
Problems and Possibilities: Strategically Pursuing a Science of Learning in Distributed Environments Stephen M. Fiore and Eduardo Solas
About the Editors
Alan R. S. Ashworth III, PhD, Air Force Research Laboratory, Brooks Air Force Base, TX Bradford S. Bell, PhD, Cornell University, Ithaca, NY Brian Bell, New Jersey Health Care System and Cognitive Engineering Research Institute, Mesa, AZ Kenneth G. Brown, PhD, The University of Iowa, Iowa City Scott R. Chaiken, Air Force Research Laboratory, Brooks Air Force Base, TX Nancy J. Cooke, PhD, Arizona State University and Cognitive Engineering Research Institute, Mesa, AZ Stephen M. Fiore, PhD, University of Central Florida, Orlando Barry P. Goettl, Air Force Research Laboratory, Brooks Air Force Base, TX Jamie C. Gorman, PhD, New Mexico State University and Cognitive Engineering Research Institute, Mesa, AZ Arthur C. Graesser, PhD, University of Memphis, TN Eddie Jerden, Development Dimensions International, Bridgeville, PA David H. Jonassen, PhD, University of Missouri, Columbia Joan Johnston, NAVAIR, Orlando, FL Steve W. J. Kozlowski, PhD, Michigan State University, East Lansing Kurt Kraiger, PhD, Colorado State University, Fort Collins Richard E. Mayer, PhD, University of California, Santa Barbara Rudy McDaniel, PhD, University of Central Florida, Orlando Harry Pedersen, New Mexico State University and Cognitive Engineering Research Institute, Mesa, AZ Eduardo Salas, PhD, University of Central Florida, Orlando Mark E. Van Buren, Corporate Executive Board, Washington, DC Robert A. Wisher, U.S. Department of Defense, Arlington, VA
The creation of this volume was partially supported by Grant F496200110214 from the Air Force Office of Scientific Research, by Grant SBE0350345 from the National Science Foundation, and by Grant N000140610118 from the Office of Naval Research. Many people contributed to this volume in differing ways. First and foremost, we thank the publishers at the American Psychological Association for supporting such a multidisciplinary edited volume. Their effort through this challenging process has surely produced a superior volume, and we sincerely appreciate the support. Second, we thank the faculty and staff of the Institute for Simulation and Training at the University of Central Florida for their continued and tireless efforts in providing a culture of scholarship and service in support of multidisciplinary science. Additionally, Stephen M. Fiore thanks his many mentors who encouraged an appreciation of multidisciplinary approaches to understanding. This includes the outstanding faculty at the University of Pittsburgh's Learning Research and Development Center who instilled an appreciation of diverse scientific knowledge in pursuit of understanding the process of learning.
Toward a Science of
1 MOVING US TOWARD A SCIENCE OF DISTRIBUTED LEARNING EDUARDO SALAS AND STEPHEN M. FIORE
The world is getting smaller. This is such a trite colloquialism—one echoed in countless conversations during the 20th century to convey the ease with which the human race could traverse the globe. Although formerly used to convey the idea of both geographic and communicative travel, in the latter portion of the 20th century this phrase truly took on a different meaning. It was now possible to engage in instantaneous virtual traversal such that one could easily project a presence across the globe. Distance no longer matters as long as the information network is in place—a fact of modem life that brings us to the rationale for this volume. Government, industry, and academia are all taking advantage of this to achieve some form of superior performance whether this is manifested as gains in learning, productivity, or national defense. Through the implementation of technologies such as distance learning or training, various organizations are rapidly adopting tools and techniques
The views herein are those of the authors and do not necessarily reflect those of their affiliated organizations. This chapter was partially supported by Grant F49620-01-1-0214 from the Air Force Office of Scientific Research to Eduardo Salas, Stephen M. Fiore, and Clint A. Bowers and by Grant SBE0350345 from the National Science Foundation to Eduardo Salas and Stephen M. Fiore.
to adapt the way they do business. With this increase in use comes the concomitant question of how to best implement such technology. Answering such questions requires a pragmatic scientific approach—pragmatic in the sense that it requires scientists to recognize that they are trying to solve real world problems and scientific in the sense that it is theory based and attempts to achieve "understanding" in addition to rinding solutions. This latter point—that the power of science needs to be better focused on helping to solve societal problems (see Stokes, 1997)—is echoed in policy discussions throughout Washington, DC, and in associated meetings across the United States. In this chapter, we briefly outline the rationale for this volume and provide an overview of the questions and associated issues we set out to address within the context of distributed learning and training.
RATIONALE FOR A VOLUME ON DISTRIBUTED LEARNING AND TRAINING In today's dynamic work environment, modifications to industrial operations and rapid advances in technology have created an unparalleled demand for training. Simultaneous with this has been a rapid advance in technology for the delivery of training along with an infusion of technologies into academic settings. Because of this, traditional classroom learning and training approaches have been increasingly supplanted by distance-learning efforts. As such, within organizations, in both the military and industry, distance learning and distributed training are becoming prevalent. For the purposes of our discussion, we use the term distributed learning and training (DLT) to generally describe learning or training that takes place while the student or trainee is geographically isolated from either the instructor or his or her peers. We choose this terminology because it covers the three primary areas in which learning at a distance may occur: (a) e-learning in organizations, that is, any type of learning facilitated using network or digital tools, (b) distance learning in academia, and (c) distributed training in the military.1 Distributed Learning and Training As evidence of the ubiquity of DLT programs such as e-learning in organizations, a recent survey published by the American Society for Training and Development (ASTD) noted that approximately 95%. of the respon-
1 In chapter 2, Kozlowski and Bell describe in more detail the differing forms of DLT and the types of technologies used in these environments.
SALAS AND FIORE
dents had used some form of DLT with their current employer (Ellis, 2003). Flexibility and cost-effectiveness are argued to be driving the increasing use of such technologies, with some estimating that more than $10 billion would be spent annually in the United States alone on DLT in the beginnings of the 21st century (Moe & Blodget, 2000). Estimates vary, but a number of large organizations have claimed to have experienced tremendous savings in their training through the implementation of DLT (e.g., Dow Chemical was reported to have saved more than $30 million, see Brayton, 2001; and IBM was reported to have saved $200 million in the training of their sales force, see Evans, 2000). Within military environments, the cost savings were reported to be nearly $300,000 a year for the National Guard (see Wisher & Priest, 1998). More recently, in a review of DLT within the military, the U.S. General Accounting Office found substantial cost and time savings associated with implementation of such programs: The Army's Battle Staff Noncommissioned Officer course conversion to an Advance Distributed Learning format resulted in a $2.9 million annual cost avoidance while maintaining student performance [and] the Air Force developed CD-ROM training for hazardous material incident response for [Department of Defense] firefighters and law enforcement personnel that reportedly resulted in a significant increase of certified responders and a projected $16.6 million cost avoidance. (U.S. General Accounting Office, 2003, p. 12)
Finally, in academia, distance learning spending in the United States alone is well into the billions, with projections for future spending approaching $10 billion (Moe & Blodgett, 2000). Nonetheless, the ubiquity of such learning environments does not necessarily match their efficacy. In academia, the use of online learning has been explored from the perspective of the learner and the characteristics determining effective learning outcomes. Along these lines, researchers have begun to understand the characteristics of success and failure for students in online courses. For example, Wang and Newlin (2000) found that online activity was proportional to success in the course. Additional research suggests that self-efficacy related to course content and to technologies related to online learning is necessary for effective performance in distance learning courses (Wang & Newlin, 2002). In organizations, an additional issue to consider is the degree to which the participants actually prefer such approaches to learning. For example, some data suggest that classroom training is preferred with more than half of those surveyed by ASTD, stating that they preferred the classroom over distance learning (ASTD & The Masie Center, 2001; see also Phillips, Phillips, & Zuniga, 2000). To address this issue, organizations have implemented blended learning environments to provide both a reduction in cost
MOVING US TOWARD A SCIENCE OF DISTRIBUTED LEARNING
and trainee desire for face-to-face meeting time. Indeed, a substantial number of organizations have implemented these forms of hybrid training programs (Sparrow, 2004) involving a mix of online or technology-enhanced learning with face-to-face teaching. For example, an IBM report indicated that the company has been able to reduce its training costs significantly while enhancing learning outcomes using DLT (Mullich, 2004). Metascientific Theme Despite these differing forms of DLT implementation, the underlying component is that people and/or content are separated by time and/or distance. This produces a host of psychological, technological, and social issues surrounding learning in such environments. Although research has begun on understanding this phenomenon on differing levels, much is yet to be learned. To address this problem, we conceptualized this volume within the scientific framework presented by Stokes (1997). In his book on basic science and technological innovation, Stokes discussed science policy from the vantage point of the history of science and humankind's quest for understanding. He described how current science policy, largely shaped by Vannevar Bush's (1945) report, "Science, the Endless Frontier," is increasingly at odds with the needs of society. In this treatise, Stokes outlined the tension that inherently arises from dichotomizing basic and applied research, a dichotomy codified in the Bush report. Stokes eloquently argued that the quest for fundamental understanding, which is historically the purview of basic science, and the consideration for use, traditionally the domain of applied research, are not mutually exclusive categories of science. Instead, when viewed from an historical perspective, these approaches have a long and productive commingling in a variety of research domains. Using Louis Pasteur's groundbreaking and significant studies in microbiology as the quintessential example of what he labeled "use-inspired basic research," Stokes outlined how science and science policy can benefit from considering research not along a single continuum of basic versus applied but within a matrix crossing a quest for fundamental understanding with a consideration of use (see also Fiore, Rubinstein, & Jentsch, 2004). We provide this introduction to Stokes's (1997) work because it is a cogent means with which to describe our approach in this volume. In particular, the domains covered in DLT exemplify the tension that exists between the basic-applied research dichotomy. On the one hand is the view that to truly understand learning and learning processes, only pure, basic research can be used to disentangle the multitude of issues surrounding such a fundamental scientific issue. On the other hand is the view that DLT is merely a new method of information delivery, and thus pure, applied research is all that is necessary to ensure gains in educational efficacy.
SALAS AND FIORE
Indeed, some argue that this latter view has dominated DLT research, and too often at the expense of developing a true science of DLT (see Salas, Kosarzycki, Burke, Fiore, & Stone, 2002). We argue that DLT can best be understood by considering it under the rubric of "use-inspired basic research." Specifically, although distance technology is indeed a means of information delivery, a small but significant number of researchers have been exploring fundamental issues in learning but within the context of DLT. In the present volume, we bring together representative members of this small group of researchers who pursue a quest for fundamental understanding, but with a well-specified consideration for use. These researchers have been exploring DLT issues from a theoretically driven perspective while simultaneously considering the eventual use of this knowledge (e.g., educational or industry applications). In short, although this technology offers tremendous promise, and is already being widely used in academia, industry, and the government, the scientific and pedagogical implications of such instruction are unclear. As such, our overall goal with this volume is to begin to truly shape a science of distributed learning that has both theoretical and practical value.
STRUCTURE OF VOLUME There are a number of areas in which DLT is making an impact: in industry, academia, and the military. As discussed, colleges and universities are increasingly offering online or blended learning environments, and public and private organizations are just as likely to offer distance training. Similarly, the military is steadily transitioning its training to online settings as well as implementing sophisticated distributed simulations for enhanced training. Thus, the issue of learning and training at a distance cutting across these domains requires that we address a multitude of differing questions, ranging from the macro to the micro. Rather than focusing on one at the expense of the other, we chose rather to present a representative cross-section of these questions to convey some of the current and influential ideas affecting DLT. Furthermore, after reflecting on cognitive psychology, technology, training, and industrial-organizational psychology's approach to use-inspired basic research, we have noticed an important increase in attempts to blend these issues. As such, we wanted to provide a focused avenue through which to present these questions and provide the beginnings of answers. Thus, in conceptualizing this volume, we set out to bring together researchers who would be able to help us begin to chart the course for a science of learning in distributed environments. Within this broader context we set out to consider a set of questions enabling an understanding of learning in distributed environments. In MOVING US TOWARD A SCIENCE OF DISTRIBUTED LEARNING
essence, the field needs to better understand why DLT may or may not be superior to co-located learning and training. What pedagogy can we adapt, and what pedagogy must we create, for these new learning environments? Both general and specific questions were generated for this volume, and these were developed to help the field move forward in addressing some of the broader issues and problems within DLT. First, more generally, we considered why there is still such a focus on technology in this area. As Mayer (1999) noted, technologies have always been introduced to the learning environment, typically along with exaggerated claims as to how learning will be greatly enhanced. As such, in conceptualizing this volume we thought at a more global level in consideration of the learner-centered approach (Mayer, 2001) and chose to present a set of what we believe are promising ideas in DLT. We chose these approaches not because they address technology in the learning environment but because of their attention to the learner's cognitive processes when interacting in technology-based learning environments. The goal is to begin laying the foundation for the principled application of technologies in DLT. Second, and still generally, we asked, Where are the broad-based theories that can help us manage the tremendous complexity of distributed learning environments? All of the contributors were asked to discuss the underlying theoretical principles driving their research. Third, and more specifically, we asked, Where is the research producing our understanding of knowledge in DLT? In industrial-organizational psychology, discussions of knowledge, skills, and attitudes are ubiquitous, yet invariably it is the skills and the attitudes that are discussed. Therefore, we included some focus on the nature of knowledge, writ large, potentially derived in DLT environments (e.g., knowledge surrounding teamwork, knowledge associated with problem solving). Fourth, and related to our third question, we asked, Where are the cognitive principles in DLT? To address this, we invited contributions from researchers who have been researching and/or developing important areas of cognition within the context of DLT (e.g., metacognition, narrative). Next, we briefly discuss the specific sections within this volume. Part I: Theoretical Models Emanating From Distributed Learning in Organizations In the first section of this book, we present a representative sample of the theoretical models addressing distributed learning within organizations. We have assembled a set of researchers who have effectively integrated organizational, technological, and cognitive issues. These authors discuss how constructs emerging from the cognitive and the learning sciences can help us to better understand organizational learning and training.
SALAS AND FIORE
First is a chapter by Kozlowski and Bell, who present a comprehensive framework of learning principles and concepts along with associated technological issues, all to be considered within distributed learning environments. As they illustrate, the field can greatly benefit from consideration of theory that can integrate the design, delivery, and use of DLT. Next is Brown and Van Buren's chapter, in which they discuss theoretical issues surrounding social capital within an organizational context. They suggest how to leverage the technologies inherent in networked organizations so as to help overcome some of the negative consequences arising from distributed interaction. Here, we see how this concept can be seamlessly blended with organizational learning and performance in today's increasingly networked settings. Finally, we have a chapter by Kraiger and Jerden; they articulate some of the foundational issues with respect to learner control and demonstrate important findings with a meta-analysis of training research that has examined this. Furthermore, on the basis of this analysis, they present an extended model of the potential differing forms learner control can manifest in organizational contexts. Part II: Distributed Teams and Distributed Team Training In Part II, we present a set of team training concepts and theories that are representative of the complex factors that must be addressed when considering distributed learning and interaction at the interindividual level. As such, we have team researchers discussing how the burgeoning field of team cognition (Salas & Fiore, 2004) and its associated training concepts need to be explored in distance and distributed learning environments. First is chapter 5, by Goettl, Ash worth, and Chaiken, in which they address how fidelity (i.e., faithfulness to the operational context) within distributed training environments alters the learning experience. They based their arguments around distinctions between fidelity at the cognitive level and the more prevalent notion of physical fidelity. As they discuss, although these differing forms of fidelity play an important role in distributed training, their use and the timing of their use must be based on sound learning principles. Next, in chapter 6, Fiore, Johnston, and McDaniel discuss how the narrative form, a concept emerging from many differing disciplines, can be productively used within distributed training environments. As they illustrate, distributed simulation-based exercises lend themselves to narrative analysis, suggesting that automated assessment and feedback delivery mechanisms can be developed by relying on the narrative form. Finally in chapter 7, Cooke et al. discuss how geographic distribution alters team process and performance within a complex military task setting. As they illustrate, team processes are altered in differing ways by distribution and understanding this effect is critical as distributed interaction becomes more prevalent.
MOVING US TOWARD A SCIENCE OF DISTRIBUTED LEARNING
Part III: Cognitive Processes and Products in Distributed Learning Environments In Part III, we offer a sample of what we describe as cognitive processes and products, for which we must account if we are to fully understand DLT. In this final section, the chapter authors discuss experimentation and theory so as to better integrate cognition and learning within distributed environ' ments. We chose three specific topics that are viewed as pressing in the sense that they represent areas of inquiry in which well-articulated theory can make a tremendous impact on DLT. First is chapter 8 by Mayer, who discusses the use of multimedia in learning environments, a use that is ubiquitous and almost synonymous with DLT. As Mayer outlines, such technologies can better augment the learning process by supporting the learner's attempts to construct knowledge associated with the domain being studied. Second is chapter 9 by Jonassen, who addresses a specific learning issue given its ubiquity in real world situations: problem solving and how problem-solving processes need to be taught within DLT. Third is chapter 10 by Wisher and Graesser, who discuss not just interactivity but a more specific form of interactivity, questioning. In particular, as Graesser and Wisher note, questions occur substantially less often in the classroom compared with settings such as those experienced during tutoring. The epistemological issue is how this important component of interaction can be better understood so that it can be incorporated into distributed learning environments. Concluding Chapter In our closing chapter to this volume, we discuss some of the broader science policy issues surrounding federal funding and how this volume can be used to strategically contribute to a science of learning in distributed environments. Our goal is to stimulate thinking at a higher level so the multiple communities involved in the science of learning can better conceptualize how the field can move forward using the contributions from this volume.
CLOSING REMARKS Our understanding of distance, distributed, and even blended learning (i.e., a mix of traditional and distance learning) environments is greatly overshadowed by the rapid rate at which they have been implemented. If academia, industry, and the military are spending billions of dollars annually
SALAS AND FIORE
on the design and delivery of distance learning and distributed training, then clearly the scientific community should support this movement by providing theory-based empirical research findings on this topic. Indeed, our professional responsibility is to investigate emerging issues, produce empirical findings, and offer those findings to help guide the field—not only in the principled application of DLT but also in areas still requiring scientific investigation. For that reason, our volume fits within policy notions of useinspired basic research and focuses entirely on theoretical and empirical aspects of DLT research. We have selected authors who can discuss important topic areas in distance learning theory, methods, and applications, and each chapter contributes to an overall understanding of major DLT issues. We believe that progress toward understanding what comprises learning in DLT environments requires more and better concepts, information, data, methods, and theories. We sincerely hope that the content of this volume begins to meet this need and encourages expanded research and theory development in this important area.
REFERENCES American Society for Training and Development & The Masie Center. (2001). E-learning: "If we build it, will they come.7" Alexandria, VA: American Society for Training and Development. Brayton, C. (2001, October 1). The learning curve: The fragmented e-learning industry rallies around a new business case: The value chain. Internet World. Retrieved January 30, 2002, from http://www.iw.com/magazine.php?inc= 100101/lO.Ol.Olebusinessl.html Bush, V. (1945). Science: The endless frontier. Washington, DC: United States Government Printing Office. Ellis, R. K. (2003, November 17). E-learning trends 2003. Learning Circuits. Retrieved February 20, 2005, from http://www.learningcircuits.org/2003/ nov2003/trends.htm Evans, S. (2000, May 15). Net-based training goes the distance. The Washington Post. Retrieved February 20, 2005, from http://www.washingtonpost.com/ac2/ wp-dyn?pagename=article&node=&contentId=A58362-2000Mayl2 Fiore, S. M., Rubinstein, J., & Jentsch, F. (2004). Considering science and security from a broader research perspective. International journal of Cognitive Technology,9, 40-42. Mayer, R. E. (1999). Instructional technology. In F. T. Durso, R. S. Nickerson, R. W. Schvaneveldt, S. T. Dumais, D. S. Lindsay, & M. T. H. Chi (Eds.), Handbook of applied cognition (pp. 551-569). Chichester, England: Wiley. Mayer, R. E. (2001). Multi-media learning. Cambridge, England: Cambridge University Press.
MOVING US TOWARD A SCIENCE OF DISTRIBUTED LEARNING
Moe, M. T., & Blodget, H. (2000, May 23). Corporate e-learning—Feeding hungry minds (Part 4, pp. 225-290). The knowledge web. Study conducted by Merrill Lynch & Co. Retrieved February 25, 2004, from http://internettime.com/itime group/MOE4.PDF Mullich, J. (2004). A second act for e-learning. Workforce Management, 83, 51-55. Phillips, J., Phillips, P. P., & Zuniga, L. (2000). Evaluating the effectiveness and the return on investment of e-learning. What Works Online: 2000, 2nd Quarter. Retrieved February 21, 2005, from http://www.astd.org/ASTD/Resources/ eval_roi_community/return.htm Salas, E., & Fiore, S. M. (Eds.). (2004). Team cognition: Understanding the factors that drive process and performance. Washington, DC: American Psychological Association. Salas, E., Kosarzycki, M. P., Burke, C. S., Fiore, S. M., & Stone, D. L. (2002). Emerging themes in distance learning research and practice: Some food for thought. International Journal of Management Reviews, 4, 135-153. Sparrow, S. (2004). Blended is better. Training and Development, 58, 52-55. Stokes, D. E. (1997). Pasteur's quadrant: Basic science and technological innovation. Washington, DC: Brookings Institution. U.S. General Accounting Office. (2003). United States General Accounting Office report to congressional committees: Military transformation, progress and challenges for DOD's advanced distributed learning programs, February 2003 (Report No. GAO-03-393). Washington, DC: Author. Wang, A. Y., &. Newlin, M. H. (2000). Characteristics of students who enroll and succeed in Web-based psychology classes. Journal of Educational Psychology, 92, 137-143. Wang, A. Y., & Newlin, M. H. (2002). Predictors of Web-student performance: The role of self-efficacy and reasons for taking an online class. Computers in Human Behavior, 18, 151-163. Wisher, R. A., &. Priest, A. N. (1998). Cost-effectiveness of audio teletraining for the U.S. Army National Guard. The American Journal of Distance Education, 12, 38-51.
SALAS AND FIORE
2 A THEORY-BASED APPROACH FOR DESIGNING DISTRIBUTED LEARNING SYSTEMS STEVE W. J. KOZLOWSKI AND BRADFORD S. BELL
There has been steady growth in the use of distance learning and distributed training over the past decade (Salas & Cannon-Bowers, 2001), with some estimates suggesting that nearly 80% of all companies use some form of distributed, computer-based training (Kiser, 2001). Although a variety of factors are stimulating this rapid growth in the use of distance learning and distributed training, two factors have been key in
This chapter is based on research sponsored in part by the Air Force Office of Scientific Research (Grant F49620-01-1-0283); S. W. J. Kozlowski and R. P. DeShon, principal investigators; and Battelle Scientific Services (Contract DAAH04-96-C-0086, TCN: 00156). The U.S. government is authorized to reproduce and distribute reprints for governmental purposes notwithstanding any copyright notation thereon. The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of the Air Force Research Laboratory, the Army Research Office, or the U.S. government.
the growth of what we refer to collectively as distributed learning systems (DLS). One factor has been the practical benefits—lower cost, rapid deployment, and flexibility—associated with training systems that transcend space and time and enable training anytime and anywhere. Another factor has been the advances in technology and connectivity. The penetration of computer technology into all facets of the workplace, the substantial increases in computing power and speed, and the interactivity enabled by the explosive growth of the Internet have provided a ready infrastructure for delivering distributed training. Indeed, the literature on distance learning and distributed training, both popular and academic, has been dominated by discussions concerning technological innovations and cost savings (Bell & Kozlowski, 2006; Kozlowski & Bell, 2002). As is often the case, the factors stimulating the attractiveness of DLS have a negative side effect. One consequence of the heavy emphasis on practical benefits and technology is that researchers and practitioners alike have paid far less attention to critical instructional design issues surrounding distributed learning. The purpose of such systems is to promote learning, yet the tendency is to design distributed learning around the media and supporting technologies rather than the underlying instructional goals and objectives of the training. This is not surprising given that there is currently no well-developed theoretical framework to guide training design for distributed systems (Salas, Kosarzycki, Burke, Fiore, & Stone, 2002). However, for DLS to be optimally effective, trainers and instructional designers must integrate learning models with instructional design practices (Schreiber, 1998; Welsh, Wanberg, Brown, & Simmering, 2003). It is critical, therefore, to develop a theoretical framework that can be used to guide DLS design. In the absence of such a framework, many organizations have discovered that their DLS, although practical and cost efficient, are suboptimal or even ineffective for developing critical knowledge and skills. As Hamid (2002) noted, the growing consensus is that "after the initial excitement, many elearning initiatives have fallen short of expectations" (p. 312). The purpose of this chapter is to present a theoretical framework to guide DLS design and enhance the effectiveness of distributed learning. The framework we develop provides theory-based principles for specifying DLS design to achieve specific instructional goals. In contrast to much of the extant literature in the areas of distributed training and distance learning, our theory views the identification of desired instructional goals and associated learning processes—not technology—as the point of departure in DLS design. We believe these goals and learning processes should drive DLS design because they elucidate the optimal instructional experience, clarify critical instructional features, and determine the technologies most appropriate for delivering the features.
KOZLOWSKI AND BELL
DESIGNING DISTRIBUTED LEARNING SYSTEMS Recent reviews of DLS research (Bell & Kozlowski, 2006; Welsh et al., 2003) and the DLS design process (Kozlowski & Bell, 2002), as well as the opening chapter of this book, highlight the need for a conceptual foundation that provides a solid empirical basis for the derivation of scientific principles to guide instructional design for DLS. To a substantial extent, progress in this area has been impeded by researchers' preoccupation with an important pragmatic concern: the bandwidth-cost trade-off problem. That is, much of the research is driven by an examination of the degree of bandwidth and interactivity required for distance learning or distributed training to approximate conventional instructor-led classroom training. Although we acknowledge that this is an important practical concern, its primary attention to cost and technology factors has misdirected attention away from the need for a conceptual foundation—one driven by learning processes and mechanisms, not technologies—to guide the design of DLS. The dilemma is that elements of the technology infrastructure are already in place or anticipated (e.g., the penetration of computer technology, enhanced connectivity, intranet and Internet access), the economic logic to harness the technology for training is compelling (e.g., upward of 80% of training costs go for indirect support rather than directly to training), and the push to practice is rapidly diffusing early efforts (e.g., there are many well-publicized e-learning efforts and initiatives in industry, education, and government). Thus, the availability of flexible technology, compelling economic drivers, and benchmark practices of early adopters are shaping the emerging nature of DLS. As a consequence, distance learning and distributed training programs are often driven by technology in terms of availability and cost rather than by instructional goals linked to desired cognitive and behavioral competencies. The availability of technology and cost factors drives the selection of delivery media. Instructional issues typically receive little or no attention in this selection process. In fact, Govindasamy (2002) noted that most learning technology vendors "deliberately distance themselves from pedagogical issues" (p. 288). Existing instructional content (e.g., manuals, lecturebased course materials), when available, is often simply mapped onto existing technology, a practice known as repurposing. That is, distributed learning technology choices are driven by what the organization has available, not what the training program requires. Because performance-relevant instructional goals are not the primary drivers of system design, attaining desired knowledge and skills as outcomes is more a matter of chance than intent. Moreover, there are two alternative outcomes that may be more likely. On the one hand, this technology-driven logic can yield training that is inefficient
DESIGNING DISTRIBUTED LEARNING SYSTEMS
because it invests in more advanced technology than is necessary for delivering the desired skills. On the other hand, it can yield training that is ineffective because it fails to use technology with sufficient capability or bandwidth to deliver an instructional experience that develops desired knowledge and skill competencies (Govindasamy, 2002). This is, in essence, the core of the bandwidth-cost trade-off. From our perspective, the current approach can yield only trial-and-error research and practice that attempt to map the boundaries of the trade-off unsystematically. In the long run, it is likely to be a slow and costly approach. Our position is that the best way to start to resolve this problem is to begin with a theoretical foundation that is driven by instructional goals and learning processes, not technologies. Training needs derived from the performance domain are used to identify desired instructional goals, which in turn implicate particular cognitive mechanisms and learning processes. Next, targeted cognitive mechanisms and learning processes guide the identification of instructional features that specify the type of content that should be delivered, how much immersion is desired, the necessary degree of interactivity, and how much communication bandwidth is essential. Desired instructional features then guide the selection of appropriate technologies and the design of a theoretically grounded instructional experience. Technology selection is appropriately located at the endpoint of the design process as a tool to ensure the delivery of an instructional experience that has been calibrated to fit training needs and instructional targets. The proposed approach can, in the short term, prescriptively suggest bandwidth targets that are likely to approximate the perimeter of the bandwidth-cost tradeoff and, in the long term, can better focus research in an effort to more precisely map the trade-off curve. We believe that this approach will yield a more timely and cost-effective research agenda to enhance the design of DLS.
THEORETICAL FOUNDATION We begin by developing a model that links instructional goals, desired knowledge and skill competencies, underlying learning processes and mechanisms, and necessary instructional design foci that deliver targeted skills. The model is designed to link the complexity of the instructional goal (basic to advanced knowledge and skills) to the types of instructional characteristics necessary to stimulate underlying cognitive-behavioral mechanisms to achieve the targeted instructional outcomes. It is important to note that because instructional goals and associated knowledge and skill competencies are sequenced from basic to advanced, higher level competencies sub-
KOZLOWSKl AND BELL
sume more fundamental knowledge and associated learning mechanisms. This model and its conceptual linkages form the theoretical core of our approach. In the second step, we develop a typology that identifies categories of instructional features that contribute to different aspects of the design of an instructional experience. This approach is different than the more typical focus on technologies and the delivery features they possess (e.g., Noe, 1999). Our focus is not on technologies per se, although different technologies—singly or in combination—are implicated by these instructional features. Instead, the idea is to first focus on the key features that enable the design of an instructional experience. The selection of technologies becomes relevant later, when the DLS infrastructure is constructed. The central conceptual characteristic distinguishing instructional features within categories is the information-experience richness of the instructional experience (low to high). As a general rule, greater information-experience richness necessitates wider bandwidth and greater cost. Next, we integrate the theoretical core and the instructional feature typology. This integration provides a basis for prescriptive guidance to predict how information richness along particular instructional feature dimensions is linked to accomplishing desired instructional outcomes at different levels of knowledge and skill complexity. Thus, this integrative typology links our theoretical core to desired instructional features, thereby providing a theoretical foundation for the DLS design. Finally, we map instructional features to discrete technologies to help guide the technology selection process. Although there are many ways of delivering different instructional features, we highlight several examples of how specific technologies can be used to deliver instructional experiences of varying levels of information/experience richness. Our goal in this final section is not to provide a complete mapping of features to all available DLS technologies but, rather, to provide several illustrative examples of the capability of specific technologies to deliver a feature-rich instructional experience. This logic underlying this mapping process can be extended beyond our examples to ensure that the technology selection for the delivery of an instructional experience is a consequence of theory-driven instructional design. We believe that by calibrating DLS delivery technologies to fit instructional goals and learning requirements, our approach will deliver cost-efficient yet effective training experiences. Moreover, we also believe that by helping to identify pragmatic concerns that are least likely to be in doubt and areas that necessitate further research, our approach will guide research toward the derivation and validation of principles for guiding the design of DLS. We now develop the rationale of this approach in more detail.
DESIGNING DISTRIBUTED LEARNING SYSTEMS
Basic Information/ Experience Richness:
Knowledge & Skill Competency:
Declarative Knowledge and Skill
Facts, concepts, rules; definition, meaning (What)
(a) Knowledge and Skill Complexity
Knowledge and Skill
Task principles; rule application (How)
Strategic Knowledge and Skill
Adaptive Knowledge and Skill
Task contingencies; selective application (When, where, why)
Generalization of task rules, principles, & contingencies (What now, what next)
Instructional Design Foci: Experi
(cilnieeration (b) DLS Features: Content: • Tex!
• Still images/graphics • Images in motion • Sound: voice, music, special effects
• Psychological fidelity • Constructive forces • Stimulus space or scope • Fidelity of context/operation: • Motion and action • Realtime • Adaptive to trainee Interactivity: • Single participants • Individual oricnied • Multiple participants High
• Team oriented
Audio and video
© S. W. J. Kozlowski & B- S. Bel], 2002, 2006, All rights reserved Used with permission.
Figure 2.1. A model linking instructional objectives, trainee competencies, and instructional features. From Enhancing the Effectiveness of Distance Learning and Distributed Training: A Theoretical Framework for the Design of Remote Learning Systems (p. 44), by S. W. J. Kozlowski and B. S. Bell, 2002. Copyright 2002 by S. W. J. Kozlowski and B. S. Bell. Reprinted with permission.
Instructional Goals, Competencies, Learning Processes, and Instructional Design Foci This model within the overall framework forms the theoretical core of our approach. As shown in the uppermost section of Figure 2.1, labeled (a) Knowledge and Skill Complexity, its foundation is provided by instructional
KOZLOWSKI AND BELL
goafs that are sequenced on the horizontal dimension from basic to advanced knowledge and skill complexity. This developmental sequence is in confermance with contemporary theories of skill acquisition that postulate the progressive development of knowledge and skill competencies, from facts to principles to contingencies to generalization (Anderson, 1982; Ford & Kraiger, 1995; Kozlowski, Toney, et al., 2001). Learning processes for acquiring this progression of competencies differ qualitatively such that the acquisition of basic knowledge necessitates encoding and is memory intensive, whereas advanced knowledge and skill acquisition requires higher level selfregulatory and metacognitive processes with an emphasis on integrating cognitive and behavioral skill. As a result, the differing learning processes that underlie different instructional goals along the developmental continuum implicate different instructional design foci. For example, at the basic end of the continuum, instructional design needs to stimulate rehearsal strategies and memorization under consistent practice conditions, whereas at the advanced end of the continuum, instruction needs to stimulate exploration and experimentation; variability in instructional stimuli and responses; and active, controlled reflection during practice. The most basic instructional goal in the sequence is the acquisition of declarative knowledge, which entails basic domain content. It focuses on knowledge of the definition and meaning of important domain facts, concepts, and rules (VanLehn, 1996). It represents basic domain content or knowledge of "what." Learning this competency involves repeated exposure to the relevant content, effortful attempts to encode the material into memory, and evaluation of current competency through tests of recall and retention as learning indicators and aids. Through practice and experience, declarative knowledge begins to be proceduralized (Anderson, 1982; Ohlsson, 1987). Procedural knowledge represents understanding of "how." This process occurs with continued practice (repetition) beyond initial successes at reproducing certain behaviors. In addition, knowledge about situations, responses, and outcomes is integrated into the knowledge to form context-specific rules for application (Ford & Kraiger, 1995; Glaser, 1994). Proceduralized knowledge, therefore, can be described as a set of condition-action rules, such as "If Condition A, then Action B" (Anderson, 1983). With experience, condition-action rules are compiled or chunked together. As a result, as knowledge is proceduralized and compiled, individuals not only are better able to determine when knowledge is applicable but also are able to apply what they know more automatically and efficiently. As knowledge and procedures continue to be compiled, more elegant task strategies emerge. Cognitive resources are freed by the internalization of behaviors, and those resources can be devoted to strategy development and self-regulation of action initiation and performance. As individuals
DESIGNING DISTRIBUTED LEARNING SYSTEMS
develop strategies and a better understanding of task situations, they integrate this knowledge into more complex mental models. The mental models of experts contain diagnostic clues for detecting meaningful patterns in the learning or transfer environment (Glaser, 1989; Kraiger, Salas, & CannonBowers, 1995). These richly interconnected knowledge structures allow experts to determine when, where, and why their knowledge applies. That is, they understand the conditions, timing, and rationale that yield effective task performance. Development of strategic knowledge requires variable practice and experimentation so that individuals can develop a complex network of structural relationships among important task concepts in the domain (Bell & Kozlowski, 2002; Kozlowski, Toney, et al., 2001). Finally, the most advanced instructional goal is the development of adaptive knowledge and skills, which are closely tied to the development of strategic knowledge and skills. However, whereas strategies are used to react to changing circumstances in a particular task context, adaptability involves generalizing and extrapolating knowledge to novel situations or tasks (Hatano & Inagaki, 1986; Holyoak, 1991). It represents knowledge of what is happening now and what one should change next to resolve the new problem or situation. Adaptability requires metacognition, which involves processes such as analyzing the situation, monitoring and evaluating one's learning progress, and controlling how to allocate one's resources and the prioritization of activities (Flavell, 1979; Schmidt & Ford, 2003). Metacognition enables an individual to recognize not only when a situation has changed but also when to discontinue a problem-solving strategy that would ultimately prove unsuccessful (Larkin, 1983). Adaptability involves dynamic replanning and the ability to pull together task-relevant knowledge to create an innovative, creative, and effective task approach. The development of adaptive knowledge and skills typically occurs well into the knowledge and skill acquisition process (Kozlowski, Toney, et al., 2001). Individuals must have relatively complete mental models of the knowledge domain, and these mental models need to contain not only valid causal relationships but also error information that is gathered from variable practice and experimentation. Existing knowledge and behaviors must also be internalized because the greater the internalization, the more cognitive resources are available for executive (metacognitive) functions (Kanfer & Ackerman, 1989). To summarize the theoretical core, we assert different instructional goals involve qualitatively different learning processes, and appropriate learning processes have to be stimulated by the DLS to accomplish the instructional goal. This provides the basic logic to guide DLS design. A fundamental aspect of the sequence of instructional goals is that complex competencies build on the foundation of basic competencies. From a training systems perspective, one key implication of this aspect of the model is that training for advanced instructional goals must ensure that (a) trainees possess
KOZLOWSKI AND BELL
basic domain competencies or that (b) provision has been made to target more basic competencies prior to targeting advanced ones. Distributed Learning System Features As we have noted, the literature on distance learning and distributed training has been dominated by a focus on technologies, with technologies then driving the form of distributed instruction. Table 2.1 catalogues the broad range of technologies that have the potential to be used in distributed learning and describes typical applications and examples. Although this approach can provide useful interventions, instructional design is ad hoc, and the technology is not tailored to deliver an instructional experience linked to a model of learning. The instructional foci derived from the model shown in Table 2.1 have the potential to be delivered in myriad ways by means of different technologies or sets of technologies. Thus, the purpose here is to look past the technologies per se and to focus instead on the kinds of instructional features—embedded in the technologies—that can be used to stimulate targeted instructional foci, thereby shaping the learning process. Our typology of (b) DLS Features, appears on the left-hand side of Figure 2.1. It classifies DLS features into four primary categories that index the richness of domain content, immersion, interactivity, and communication that can be delivered by DLS. Within categories, features are organized from low to high with respect to the richness of the information or experience they can create for trainees. The first category, content, concerns the richness with which basic information (declarative knowledge) is delivered through the system to trainees. In its most sparse form, information is conveyed as text. Text is quite flexible and is a near-universal capability in most DLS. Additional features, such as still images and graphics, images in motion, and sound, can be added to basic text to enrich the information stream. It is important to recognize that more information is not necessarily richer information. Multimedia elements enhance learning only when they help the learner understand and make sense out of the material (e.g., Mayer & Anderson, 1992). The second category focuses on features that influence immersion, or sense of realism. This category concerns the extent to which the training captures key psychological characteristics of the performance domain (i.e., psychological fidelity) and, beyond that, draws trainees into the experience—that is, creates a micro- or synthetic world that captures their attention and subjects them to important contextual characteristics relevant to the performance domain (i.e., gradations of the physical fidelity of the experience; Schiflett, Elliott, Salas, & Coovert, 2004). This category is particularly important with respect to simulation design (e.g., distributed interactive simulation [DIS] and distributed mission training, DMT), because
DESIGNING DISTRIBUTED LEARNING SYSTEMS
TABLE 2.1 Distributed Learning System Technologies System
• Integrates text, graphics, animation, audio, and video. • Computer-based delivery allows trainees to interact with content by typing responses, using a joystick, or using a touch-screen monitor.
Colorado Springs Fire Department has created digital simulations of fires on DVDs. Depending on decisions that trainees make throughout the program, the fire either gets better or worsens. Dow Chemical places many of its training programs on CDROM so that employees can participate in the training where and when they want.
• Instruction is provided one on one to trainees via a monitor connected to a keyboard. • Trainees use the keyboard or touch the monitor to interact with the program.
Federal Express has an interactive video curriculum that covers courses on customer etiquette, defensive driving, and delivery procedures. Employees control the content they view and where and when they participate in the training.
• Can allow communication between trainers and trainees and among trainees. • Online referencing. • Testing assessment. • Distribution of computer-based training. « Delivery of multimedia. • Trainees can use hyperlinks to interact with the program.
CIGNA uses Web-based training programs to deliver training to its nurse consultants distributed around the country, many of them in rural areas. General Electric has created Web-based environmental, safety, and health training in 17 languages to ensure compliance among its operating locations distributed around the world.
• Provides trainees with a threedimensional learning experience. • Trainees move through the simulated environment and interact with its components. • Trainees in different locations can be linked in a simulated environment.
Ford Motor Company uses virtual reality simulations to train new employees in its Vulcan Forge unit. Employees are fitted with a head-mount display that allows them to view the virtual world and they handle tools that are the same size and weight as those that they will use on the job. The virtual environment allows employees to learn the potentially hazardous job in a safe environment. (continued)
KOZLOWSKl AND BELL
TABLE 2.1 System
• Refers to instructional systems that utilize artificial intelligence to provide individualized instruction. • Trainee performance is analyzed to provide feedback and coaching and also to generate future scenarios and instruction.
The U.S. Navy uses an intelligent tutoring system in its officer tactical training. The system uses simulation and provides an automated evaluation of each student's actions. The system has enabled the program to be offered as self-study, and learners receive 10 times more hands-on experience than before.
Electronic • Computer applications that properformance vide skills training, information support access, or expert advice upon systems request. • Often used as an employee assistance device, but can also be used as a training tool.
American Express uses an Electronic Performance Support System to train its customer service accounts staff. The system helps employees to deal with problems by structuring information, coaching them, prompting for required information, and serving as a reference for information on company products and policies.
Distributed • interactive simulations • (DISs), synthetic task environments (STEs)
Low-fidelity, PC-based Simulations of real world tasks, Linked by a standard interaclive protocol, dispersed Simulations can be linked together to emulate teams and teams of teams performing in a realtime virtual environment or microworld.
The U.S. military uses a variety of DIS and STE platforms to model team performance and conduct research on team effectiveness research (ODD, TEAM/ Sim, AEDGE). As the psychological fidelity of these simulations improves, DISs and STEs are increasingly being used as basic training tools for military team effectiveness.
• High-fidelity simulations of real world tasks, replicating operational equipment, procedures, and task demands. • Both "real" and constructed or virtual entities create a complex and rich performance context. • Linked by a standard interactive protocol, simulations located worldwide can be linked together to emulate teams and teams of teams performing in a virtual task environment.
The U.S. military has experimented with a variety of DMT systems to gain experience and lessons learned for the development of this training technology. Examples include ROADRUNNER '98, SIMNET, and JEX. By allowing trainees to participate in a common, but complex and diverse battlespace, DMT systems allow the honing of high level skills, safely, and at low cost.
Intelligent tutoring systems
Distributed mission training (DMT)
Note. From Enhancing the Effectiveness of Distance Learning and Distributed Training: a Theoretical Framework for the Design of Remote Learning Systems (Final Report; Contract No. DAAH04-96-C-0086, TCN: 00156), by S. W. J. Kozlowski and B. S. Bell, 2002, Research Triangle Park, NC: Battelle Scientific Services. Copyright 2002 by Battelle Scientific Services. Reprinted with permission.
DESIGNING DISTRIBUTED LEARNING SYSTEMS
it provides a basis to identify features that help scale the immersion potential of lower to higher fidelity synthetic task environments (STEs) and task scenarios. The presumption is that psychological fidelity in terms of central constructs, processes, and performance measures provides an essential basic foundation for learning and that gradations of physical fidelity add contextual realism that further grounds the instructional experience to important cues and contingencies present in the performance domain (Kozlowski & DeShon, 2004). Up to this point, the two categories of features that have been discussed are common to all instructional systems and represent important instructional design choices that shape the instructional experience. The next two categories, however, are unique to DLS in that they enable distribution of instruction by means of communication media, and they shape the nature of the distributed instructional environment by determining the type of interaction that can ensue. They are the features that make the learning experience "distributed." The category labeled interactivity considers characteristics that can influence the potential degree and type of interaction between remote instructors and students; among distributed student peers; and, potentially, among multiply distributed student teams or collaborative learning groups (e.g., Bouras, Philopoulos, & Tsiatsos, 2001; Collis & Smith, 1997). The potential range of interactivity is dependent on communication bandwidth, which is considered as a separate category. However, the interactivity category presents a set of design options in its own right. Given that instructors and students are distributed in space (and potentially in time), the issue here concerns the degree to which learning is centered on the individual learner in relative isolation, or whether learners are to be linked into clusters to enable teamwork or collaborative learning. There is a growing theory and research literature indicating that training for teamwork skills—coordination, communication, and adaptability— necessitates a team context (Cannon-Bowers, Tannenbaum, Salas, & Volpe, 1995; Kozlowski, 1998; Kozlowski & Bell, 2003). Thus, instructional design for some applications of distributed learning—especially those using STEs such as DMT and DIS involving teams and teams of teams—must incorporate considerations of team interaction as it relates to training desired performance competencies. In addition, there is an emerging literature on collaborative learning suggesting that appropriate instructional supports can help learners to teach each another and that learners can learn more and better under collaborative learning conditions (e.g., O'Donnell, 1996; Rosenshine & Meister, 1995). Given the distance or potential absence of instructors, collaborative learning principles may have the potential to augment instruction and to supplement instructor guidance for DLS. How such principles could be applied would be constrained by whether students were independently distributed—necessitating remote collaboration—or
KOZLOWSKI AND BELL
whether they were distributed in cluster sites, allowing face-to-face collaborative learning. The last category, features that influence communication richness, concerns the issue of communication bandwidth. Conventional face-to-face instruction places experienced instructors and students in spatial and temporal proximity. This enables the expert instructor to evaluate student learning in real time by monitoring student reaction cues (e.g., nodding vs. a puzzled face) and testing for comprehension (e.g., asking a probing question). It also allows students to share views, perspectives, and comprehension with each other. However, when students are distributed in space (and potentially in time), real-time access and processing of such latent communication cues are dependent on the bandwidth of the communication link (Guzley, Avanzino, & Bor, 2001). Also, even at its highest bandwidth (e.g., synchronous, real-time, audio-video link), latent cues and strategies for managing instruction, conversation, and exchange are degraded: Field of view is reduced, the ability to gesture is limited, facial expressions are eliminated or constrained, auditory cues are diminished, tools and artifacts are difficult to share, and shared information is delayed or decoupled from its context. On the one hand, these concerns regarding communication richness have formed the primary focus of research evaluating the effectiveness of distance learning relative to conventional face-to-face classroom instruction (e.g., Faux & Black-Hughes, 2000; Huff, 2000; Meisel & Marx, 1999; Wisher &. Curnow, 1999). The essential question is, How much communication bandwidth is necessary to approximate the same instructional quality across environments? On the other hand, other literatures concerned with remote collaboration or computer-mediated communication suggest some positive aspects of lower information richness (e.g., asynchronous, time delayed, text only) that may enhance information exchange, at least for some individuals. For example, status differentials are reduced, responses can be more thoughtful, and introverted or culturally dissimilar individuals may be more likely to participate (see McGrath & Hollingshead, 1994; McKenna & Green, 2002). In addition, communication can entail more than person-to-person conversation, and production blocking can be eliminated. DLS will often necessitate information and data exchange exclusive of conversation. Thus, any way one considers the issue, communication bandwidth is an important consideration in the design of DLS. Integrating Instructional Objectives, Competencies, and Features Implications for DLS design that are provided by the integration of these two conceptual models are shown in the core of Figure 2.1 and labeled (c) Integration. Each DLS feature category is associated with a rectangular area demarked into two zones that correspond to the applicability of the
DESIGNING DISTRIBUTED LEARNING SYSTEMS
feature(s) on the vertical dimension of information/experience richness to the targeted instructional goal, competency, and instructional design foci along the horizontal dimension. The key conceptual contribution is the diagonal demarking the two zones that posits the applicability of the instructional features to targeted instructional goals. The diagonal signifies the hypothe' sized degree of information-experience richness (i.e., bandwidth trade-off) required to achieve targeted instructional objectives. The white zone corresponds to features that are essential to meeting the desired instructional goal. The shaded zone corresponds to features that are optional, at the cost of additional technology and bandwidth. Although additional features may augment the instructional experience, they may not be necessary and may yield cost inefficiencies. For example, when the targeted instructional goal is declarative knowledge, the diagonal references text as the primary content, psychological fidelity as the key immersion feature, single participants as the interactivity target, and one-way communication as the enabling link. In other words, the leading edge of the diagonal maps the most parsimonious, instructionally effective, and cost-efficient features that should drive the specification of delivery technologies for the targeted instructional objective. Additional features—at the cost of greater bandwidth and more advanced technological infrastructure—may augment the instructional experience, but the tradeoff of cost efficiency relative to any increment in instructional effectiveness is an open question. Moving away from the leading edge of the diagonal into the shaded area suggests the potential degree of inefficiency. In contrast, when the targeted instructional goal is adaptive knowledge and skill, the leading edge of the diagonal references the highest degree of informationexperience richness (and the subsumed lower level features) to accomplish the objective. That is, failure to use sufficient information—experience richness will likely yield ineffectiveness relative to achieving the targeted instructional goal. The purpose of this model is to specify how instructional goals can be used to guide the selection of essential instructional features needed to provide an instructional experience of sufficient richness to achieve targeted objectives. Once the necessary instructional features have been identified, the next step is to select a technology system or infrastructure that can deliver the desired level of information-experience richness. We discuss this final stage of the process in the next section. The diagonal in Figure 2.1 represents the hypothesized bandwidth trade-off. It is likely that—rather than a linear diagonal—the boundary or leading edge defining the essential and optional features is a nonlinear curve. Precise specification of this curve will necessitate targeted research to test and map the boundary. However, in the interim the model can serve as a prescriptive and predictive tool for specifying cost-efficient and effective instructional features and delivery
KOZLOWSKl AND BELL
technologies for achieving particular instructional goals. We now turn our attention to the process of selecting a delivery technology that can provide the necessary instructional features. Linking Instructional Features to Distributed Learning System Technologies The final stage of the DLS design process involves mapping the necessary instructional features against potential delivery technologies. All too often, the selection of DLS technologies has served as the first step in the design process. The theory-based approach we advocate, however, views technology selection as an activity that concludes the DLS design process. Technology simply serves as the medium by which to deliver the instructional experience necessary to stimulate critical learning processes and develop targeted knowledge and skills. Thus, competent technology selection can occur only when one has first specified the goals of instruction and identified the level of information richness necessary on each of the four features to achieve those goals. Examples of specific technologies that can be used to achieve different levels of information—experience richness on the four distributed learning features are provided in Table 2.2. This goal of the table is not to present a comprehensive mapping of instructional features to DLS technologies because technological combinations or variants create the potential for a vast number of unique DLS applications that can have very different capabilities. Instead, the table is designed to illustrate the level of information richness that can be achieved on each of the four critical instructional features using different types of technologies. We also highlight the specific technological features that implicate the ability of these systems to offer different levels of information richness. If one focuses on content, one sees that the lowest level of information richness is provided by printed material, which contains only text and images. As one moves up the information richness continuum, there is Web-based text, which has the potential to include motion, and then at the highest levels there is interactive, live instruction, which combines text, images, motion, video, sound, and special effects. Because the technology uses a multimedia format there is an opportunity to present information so that it targets multiple sensory modalities (e.g., visual, verbal; Clark & Mayer, 2002; Mayer, 2001). Presenting content through multiple sensory channels can enhance learning, although it is important to avoid overload because learners have a limited cognitive capacity (Mayer, 2001; Paas, Renkl, & Sweller, 2003). Video-based programs provide the lowest level of immersion, offering basic psychological and physical fidelity. Immersion is enhanced by programs
DESIGNING DISTRIBUTED LEARNING SYSTEMS
TABLE 2.2 Illustrative Examples Linking Distributed Learning System Features to Specific Technologies Distributed learning features Level of information/ experience richness
Video (basic psychological and physical fidelity)
CD-ROM (individual oriented)
Web-based text presentation (1-way, asynchronous communication)
Web-based text •presentation (text, images, motion)
Web-based interactive media program (psychological fidelity, potential for humancomputer interaction)
Web-based program with group support systems (trainee interaction)
Online learning communities (text-based, 2-way, synchronous communication)
Interactive, live instruction (text, images, motion, video, sound, special effects)
Virtual reality (high psychological fidelity, motion, action, adaptive)
Distributed interactive simulation (multiple participant/ team oriented)
Video conferencing (video-based, 2-way, synchronous communication)
Printed material (text, images)
such as Web-based interactive media programs, whicb create interactive content, but there is a significant jump in the level of immersion offered by virtual reality programs. Virtual reality offers a three-dimensional representation of the environment and provides an opportunity for realistic human-human or human-machine interactions. The result is high levels of both psychological and physical fidelity. With respect to interactivity, at the high end of the continuum is DIS, which is a PC-based, networked simulation that allows individuals to participate in hands-on exercises with other participants or teams. The programs typically simulate real-world environments and allow real-time communications and interaction, although the level of immersion is typically low because of a two-dimensional representation of the environment (Schiflett et al., 2004). In Web-based environments, moderate levels of interactivity can be achieved by incorporating group support systems—such as chat, bulletin boards, or Webcams—
KOZLOWSKl AND BELL
that increase the level of interactivity over individually oriented programs, such as CD-ROM programs. Many Web-based programs use one-way, asynchronous communication because information is being provided to the learner, but the learner does not have an opportunity to communicate with the instructional system or other learners. To provide an opportunity for social learning in Web-based environments one can use online learning communities that allow learners to communicate with one another through text-based messaging (Johnson & Huff, 2000). When social learning is critical for achieving desired instructional goals, as it often is in team-based learning environments, more information-rich communication technologies, such as video conferencing, can be used. Video conferencing offers not only an opportunity for learners to communicate verbally but also an opportunity to send visual signals, which may be important when visual cues are critical to learning or task performance. Although the previous examples consider the distributed learning features individually, it is important to recognize that these features combine to create an instructional experience, and it is likely that the level of information richness required on one feature is not entirely independent of that required on other features. For example, if a high level of immersion is necessary for learning, then the content will likely need to be presented through multiple sensory modalities. Similarly, if a high level of interactivity is desired, then it is also likely that two-way, synchronous communication systems will need to be used. Thus, it is important to consider these interconnections and to use a technology that can deliver the level of information richness necessary in all categories. This may necessitate the blending of different technologies to gain access to necessary features.
CONCLUSION AND IMPLICATIONS The growing use of distributed, technology-based training systems by organizations is both exciting and potentially problematic. Recent advances in technology have facilitated the development of a host of new and innovative training tools, such as virtual reality and interactive media, and have allowed organizations to devise training programs that transcend boundaries of space and time. At the same time, however, both research and practice surrounding DLS have been driven largely by pragmatic concerns, such as the bandwidth-cost trade-off, and many critical instructional issues surrounding DLS have received little or no attention. In effect, the availability of flexible technology, compelling economic drivers, and benchmark practices of early adopters have shaped the emerging nature of DLS. As a consequence, there
DESIGNING DISTRIBUTED LEARNING SYSTEMS
currently exists the potential for organizations to develop DLS that are ineffective for developing employee knowledge and skills. At best, organizations are likely to underutilize the instructional capabilities of distributed learning and, therefore, greatly limit their potential return on investment. We have addressed this problem in this chapter by developing a theoretical framework that can be used to guide the design of DLS. In contrast to most of the extant literature, the framework is driven by instructional goals and learning processes, not technologies. We argue that DLS design should begin with the identification of desired instructional goals to identify specific cognitive mechanisms and learning processes. Targeted cognitive mechanisms and learning processes then guide the identification of the instructional features and content necessary to stimulate knowledge and skill development. Finally, desired instructional features guide the selection of appropriate technologies and the design of a theoretically grounded instructional experience. We believe this approach to DLS design will ensure the delivery of an instructional experience that has been calibrated to fit training needs and instructional targets, thereby producing a more effective and efficient DLS. Implications for Future Research and Practice Research Implications
Given the limitations associated with the logic that currently drives DLS design, we believe the theoretical framework described in this chapter provides instructional designers and trainers with a valuable prescriptive tool. The theoretical framework provides a conceptual foundation for principles that can provide guidance at critical stages of the DLS design process. Nevertheless, the framework is preliminary, and it is important to conduct research to validate, evaluate, and refine it. Empirical work is needed to validate the proposed linkages between instructional goals, instructional design foci, and technology features; to evaluate how the fit between these conceptual dimensions influences DLS effectiveness; and to refine the framework and its underlying principles. One potential area of research focus in this regard is the leading edge demarking the boundary between essential and optional instructional features given a particular instructional goal and associated learning processes illustrated in Figure 2.1. Although the conceptual mapping represents the boundary as a linear diagonal, differences in information richness across the DLS features are not likely to exhibit smooth linearity in practice. Instead, it is more likely that specific DLS features will vary in their incremental contribution to information richness; some will add much, whereas others may add relatively little at any particular point along the knowledge and skill complexity continuum. Thus, the leading edge is more likely to be represented by a curve, one with discontinuities for particular DLS features.
KOZLOWSKl AND BELL
We believe that in the absence of precise mapping data, the linear assumption inherent in the model is reasonable. However, the leading edge represents the cost-bandwidth trade-off and, to the extent to which one is interested in maximizing impact while minimizing cost, the boundary needs to be more precisely mapped for different DLS features. Inherent in this process is the need for research to provide a more precise evaluation and cataloguing of the instructional information—experience richness of different technologies. Research is also needed to explore how information richness may interface with other characteristics that define the instructional experience. For example, some researchers have focused attention on issues such as cognitive load or information processing (Clark & Mayer, 2002; Mayer, 2001; Paas et al., 2003). It is likely that information richness has implications for information processing and the cognitive load of trainees in that greater information richness is likely to necessitate more information processing, placing a larger cognitive load on the trainee (Kalyuga, Chandler, &. Sweller, 1999). Researchers should also examine how different strategies, such as segmenting (allowing time between segments of material) or synchronizing (present corresponding visual and verbal information simultaneously), can be used to reduce cognitive load in information-rich learning environments (e.g., Kalyuga et al., 1999; Mayer & Moreno, 2003). In the interim, however, we believe the model is a valuable prescriptive tool for specifying cost-efficient and effective instructional features and delivery technologies for achieving particular instructional goals. The theoretical framework we have developed can, in the short term, help guide distributed learning practice and, in the long term, stimulate a focused research agenda that will produce empirically grounded scientific principles to guide DLS design and ensure DLS effectiveness and efficiency. Research Extensions Beyond the focus of this model—integrating instructional goals, learning processes, and the design of DLS technologies—other theoretical issues are relevant to enhancing DLS design and thus are deserving of further consideration. For example, one set of issues that are particularly relevant concerns the integration of instructional supports that can be embedded in DLS design to further enhance the instructional experience. Some of these instructional supports are similar to those used in traditional learning environments (e.g., feedback), but the way these instructional supports are designed and implemented may be influenced by the unique characteristics of distributed learning environments. Hamid (2002), noted, for example, In an e-learning situation, a student is prone to frustration because of the technical skills required, the isolation, and because an online class lacks the built-in conventions. User frustration can be minimized
DESIGNING DISTRIBUTED LEARNING SYSTEMS
through embedding support and feedback features such as chat rooms, active links, and perhaps by providing a time management system, (pp. 314-315)
Parush, Hamm, and Shtub (2002) considered metacognitive issues in the context of distributed learning, suggesting that it is important to build in functions that can support or enhance metacognitive activities. They added a learning history function in their study that facilitated learners' ability to review and evaluate past performance. The issue of metacognition in synthetic task environments also was highlighted by Fiore, Cuevas, Scielzo, and Salas (2002), who focused on the importance of metacognition for training individuals for distributed mission teams. As another example of an instructional support strategy, work by Paas (e.g., Paas & Van Merrienboer, 1994) has considered the use of "worked examples" in computer-based training to reduce the level of cognitive load experienced by trainees. Our own work in this area of providing instructional supports in technology-based training has focused on learning processes and how the focus and quality of self-regulation can be leveraged by means of instructional design and supports (Kozlowski, Toney, et al., 2001) to prompt active learning (Smith, Ford, & Kozlowski, 1997). Some of the more effective active learning tools include the use of mastery-oriented versus performanceoriented goals to prompt more effective learning and adaptation (Kozlowski, Gully, et al., 2001); prospective adaptive guidance that guides the learner to make appropriate choices about what to study and practice given current levels of learning (Bell & Kozlowski, 2002); and the synergistic combination of instructional, motivational, and emotional control elements that prompt more effective self-regulation and learning in the open, learner-controlled environment that typifies technology-based and distributed training (Bell & Kozlowski, 2003). The point is that the area of instructional supports is wide open for theory development and research, and it is an important adjunct to the basic theoretical foundation we have developed in the chapter. The next step in our systematic research effort will be to integrate it with the framework presented in this chapter. Implications for Practice
Influencing the practice of DLS design is the target of our approach, with principles to guide practice falling directly from our theoretical framework. Thus, the basic application of the model begins with the specification of desired instructional goals, targeted knowledge and skill competencies, and associated learning processes that then indicate the types of instructional foci or experiences needed to develop the targeted skills and the sets of technologies that can deliver a corresponding instructional experience.
KOZLOWSKI AND BELL
The application of this approach clearly will necessitate a multidiscipli' nary effort. Organizations or training design firms will need to use the services of instructional designers, educational psychologists, cognitive psychologists, and other specialists expert in the science of learning who are working in partnership with computer programmers, Web designers, and other technology and media specialists expert in the art of creating an experience. There is a pressing need for more cross-disciplinary interaction; achieving a balance between these design foci is critical. Finally, we believe that those developing retum-on-investment models will need to consider the effectiveness of distributed learning from an instructional standpoint. That is, to what extent does a DLS deliver targeted knowledge and skill competencies? The dominant focus on operating costs has produced what appear to be large return-on-investment estimates, but if learning is not considered then it is difficult to determine whether the program is really having the desired effect. This evolution of program evaluation is necessary to further shift attention to instructional issues in this field. Conclusion In this chapter, we have developed a theoretical framework that can be used to guide future research and practice surrounding the development, design, and use of DLS. Given the lack of existing theory and the inadequacies of current research, the framework provides a foundation for a research agenda that is focused on the development of an elaborated set of scientific principles to guide DLS design and to address important theoretical and pragmatic concerns. Although there is no question that distributed learning holds the potential to enhance training and organizational effectiveness, a theoretically grounded approach to DLS design is needed if this potential is to be realized. We hope that this chapter stimulates efforts to elaborate our approach.
REFERENCES Anderson, J. R. (1982). The acquisition of cognitive skill. Psychological Review, 89, 369-406. Anderson, J. R. (1983). The architecture of cognition. Cambridge, MA: Harvard University Press. Bell, B. S., & Kozlowski, S. W. J. (2002). Adaptive guidance: Enhancing selfregulation, knowledge, and performance in technology-based training. Personnel Psychobgy, 55, 267-307.
DESIGNING DISTRIBUTED LEARNING SYSTEMS
Bell, B. S., & Kozlowski, S. W. J. (2003, April). An examination of the instructional, motivational, and emotional elements of active learning. In B. S. Bell &. S. W. J. Kozlowski (Chairs), Active learning: Critical elements, instructional supports, and (earning processes. Symposium conducted at the 18th annual conference of the Society for Industrial and Organizational Psychology, Orlando, FL. Bell, B. S., &. Kozlowski, S. W. J. (2006). Distributed learning system design: A new approach and an agenda for future research. Center for Advanced Human Resource Studies Working Paper Series (CAHRS Working Paper Series, (No. 06-07). Ithaca, NY: CAHRS. Bouras, C., Philopoulos, A., & Tsiatsos, T. (2001). E-learning through distributed virtual environments. Journal of Network and Computer Applications, 24, 175199. Cannon-Bowers,]. A., Tannenbaum, S. I., Salas, E., & Volpe, C. E. (1995). Denning team competencies and establishing team training requirements. In R. Guzzo & E. Salas (Eds.), Team effectiveness and decision making in organizations (pp. 333-380). San Francisco: Jossey-Bass. Clark, R. C., &. Mayer, R. E. (2002). E-leaming and the science of instruction: Proven guidelines for consumers and designers of multimedia learning. San Francisco: Jossey-Bass. Collis, B., & Smith, C. (1997). Desktop multimedia environments to support collaborative distance learning. Instructional Science, 25, 433-462. Faux, T. L., & Black-Hughes, C. (2000). A comparison of the Internet versus lectures to teach social work history. Research on Social Work Practice, 10, 454-466. Fiore, S. M, Cuevas, H. M., Scielzo, S., & Salas, E. (2002). Training individuals for distributed teams: Cognitively diagnostic assessment for distributed mission research. Computers in Human Behavior, 18, 729-744. Flavell, ]. H. (1979). Metacognition and cognitive monitoring: A new area of cognitive-developmental inquiry. American Psychologist, 34, 906—911. Ford, J. K., & Kraiger, K. (1995). The application of cognitive constructs and principles to the instructional systems model of training: Implications for needs assessment, design, and transfer. In C. L. Cooper &. I. T. Robertson (Eds.), International review of industrial and organizational psychology (Vol. 10, pp. 1-48). New York: Wiley. Glaser, R. (1989). Expertise in learning: How do we think about instructional processes now that we have discovered knowledge structures? In D. Klahr & K. Kotovsky (Eds.), Complex information processing: The impact of Herbert A. Simon (pp. 269-282). Hillsdale, NJ: Erlbaum. Glaser, R. (1994). Learning theory and instruction. In G. d'Ydewalle, P. Eelen, & P. Bertelson (Eds.), International perspectives on psychological science: Vol. 2, The state of the art (pp. 341-357). Hove, England: Erlbaum. Govindasamy, T. (2002). Successful implementation of e-learning: Pedagogical considerations. The Internet and Higher Education, 4, 287-299.
KOZLOWSKI AND BELL
Guzley, R. M., Avanzino, S., & Bor, A. (2001). Simulated computer-mediated/ video-interactive distance learning: A test of motivation, interaction satisfaction, delivery, learning & perceived effectiveness, journal oj Computer--Mediated Communication, 6(3). Retrieved December, 5, 2006, from http://jcmc.indiana. edu/vol6/issue3/guzley.html Hamid, A. A. (2002). e-Learning: Is it the "e" or the learning that matters? Internet and Higher Education, 4, 311-316. Hatano, G., &. Inagaki, K. (1986). Two courses of expertise. In H. Stevenson, H. Azuma, &. K. Hatuka (Eds.), Child development and education in japan (pp. 262-272). San Francisco: Freeman. Holyoak, K. J. (1991). Symbolic connectionism: Toward third-generation theories of expertise. In K. A. Ericsson & J. Smith (Eds.), Toward a general theory of expertise: Prospects and limits (pp. 301-355). Cambridge, England: Cambridge University Press. Huff, M. T. (2000). A comparison study of live instruction versus interactive television for teaching MSW students critical thinking skills. Research on Social Work Practice, 10, 400-416. Johnson, M. M., & Huff, M. T. (2000). Students' use of computer-mediated communication in a distance education course. Research on Social Work Practice, 10, 519-532. Kalyuga, S., Chandler, P., & Sweller, J. (1999). Managing split-attention and redundancy in multimedia instruction. Applied Cognitive Psychology, 13, 351371. Kanfer, R., & Ackerman, P. L. (1989). Motivation and cognitive abilities: An integrative/aptitude-treatment interaction approach to skill acquisition. Journal of Applied Psychology, 74, 657-690. Kiser, K. (2001, October). State of the industry 2001. Minneapolis, MN: Online Learning Magazine. Retrieved February 1, 2002, from http://www.online learningmag.com/onlinelearning/images/pdf/2001state_of_industry.pdf Kozlowski, S. W. J. (1998). Training and developing adaptive teams: Theory, principles, and research. In J. A. Cannon-Bowers & E. Salas (Eds.), Decision making under stress: Implications for training and simulation (pp. 115-153). Washington, DC: American Psychological Association. Kozlowski, S. W. J., 6k Bell, B. B. (2002). Enhancing the effectiveness of distance learning and distributed training: A theoretical framework for the design of remote learning systems (Final Report, Contract No. DAAH04-96-C-0086, TCN: 00156). Research Triangle Park, NC: Battelle Scientific Services. Kozlowski, S. W. J., & Bell, B. S. (2003). Work groups and teams in organizations. In W. C. Borman, D. R. Ilgen, & R. J. Klimoski (Eds.), Handbook of psychology: Industrial and organizational psychology (Vol. 12, pp. 333-3 75). New York: Wiley. Kozlowski, S. W. J., & DeShon, R. P. (2004). A psychological fidelity approach to simulation-based training: Theory, research, and principles. In E. Salas, L. R. Elliott, S. G. Schflett, & M. D. Coovert (Eds.), Scaled worlds: Development, validation, and applications (pp. 75-99). Burlington, VT: Ashgate.
DESIGNING DISTRIBUTED LEARNING SYSTEMS
Kozlowski, S. W. ]., Gully, S. M., Brown, K. G., Salas, E., Smith, E. A., &. Nason, E. R. (2001). Effects of training goals and goal orientation traits on multidimensional training outcomes and performance adaptability. Organisational Behavior and Human Decision Processes, 85, 1-31. Kozlowski, S. W. J., Toney, R. J., Mullins, M. E., Weissbein, D. A., Brown, K. G., & Bell, B. S. (2001). Developing adaptability: A theory for the design of integrated-embedded training systems. In E. Salas (Ed.), Advances in human performance and cognitive engineering research (Vol. 1, pp. 59-123). New York: JAI Press. Kraiger, K., Salas, E., &. Cannon-Bowers, J. A. (1995). Cognitively-based measures of learning during training. Human Factors, 37, 1-13. Larkin, J. H. (1983). The role of problem representation in physics. In D. Centner & A. L. Stevens (Eds.), Mental models (pp. 75-98). Hillsdale, NJ: Erlbaum. Mayer, R. E. (2001). Multi-media learning. Cambridge, England: Cambridge University Press. Mayer, R. E., & Anderson, R. B. (1992). The instructive animation: Helping students build connections between words and pictures in multimedia learning. Journal of Educational Psychology, 84, 444-452. Mayer, R. E., & Moreno, R. (2003). Nine ways to reduce cognitive load in multimedia learning. Educational Psychologist, 38, 43-52. McGrath, J. E., & Hollingshead, A. B. (1994). Groups interacting with technology. Thousand Oaks, CA: Sage. McKenna, K. Y. A., & Green, A. S. (2002). Virtual group dynamics. Group Dynamics: Theory, Research, and Practice, 6, 116-127. Meisel, S., & Marx, B. (1999). Screen to screen versus face to face: Experiencing the difference in management education. Journal of Management Education, 23, 719-732. Noe, R. A. (1999). Employee training and development. Burr Ridge, IL: McGraw-Hill. O'Donnell, A. M. (1996). Effects of explicit incentives on scripted and unscripted cooperation, journal of Educational Psychology, 88, 74-86. Ohlsson, S. (1987). Truth versus appropriateness: Relating declarative to procedural knowledge. In D. Klahr, P. Langley, & R. Neches (Eds.), Production system models of learning and development (pp. 287-327). Cambridge, MA: MIT Press. Paas, F., Renkl, A., & Sweller, j. (2003). Cognitive load theory and instructional design: Recent developments. Educational Psychologist, 38, 1-4. Paas, F. G. W. C., & Van Merrienboer, J. J. G. (1994). Variability of worked examples and transfer of geometrical problem-solving skills: A cognitive-load approach. Journal of Educational Psychology, 86, 122-133. Parush, A., Hamm, H., & Shtub, A. (2002). Learning histories in simulation-based teaching: The effects on self-learning and transfer. Computers & Education, 39, 319-332.
KOZLOWSKI AND BELL
Rosenshine, B., &. Meister, C. (1995). Direct instruction. In L. W. Anderson (Ed.), International encyclopedia of teaching and teacher education (2nd ed., pp. 143-149). Oxford, England: Pergamon. Salas, E., & Cannon-Bowers, J. A. (2001). The science of training: A decade of progress. Annual Review of Psychology, 52, 471-499. Salas, E., Kosarzycki, M. P., Burke, C. S., Fiore, S. M., & Stone, D. L. (2002). Emerging themes in distance learning research and practice: Some food for thought. International journal of Management Reviews, 4, 135-153. Schiflett, S. G., Elliott, L. R., Salas, E., & Coovert, M. D. (Eds.). (2004). Scaled worlds: Development, validation, and application. Surrey, England: Ashgate. Schmidt, A. M., & Ford, J. K. (2003). Learning within a learner control training environment: The interactive effects of goal orientation and metacognitive instruction on learning outcomes. Personnel Psychology, 56, 405-429. Schreiber, D. A. (1998). Organizational technology and its impact on distance training. In D. A. Schreiber & Z. L. Berge (Eds.), Distance training: How innovative organizations are using technology to maximize learning and meet business objectives (pp. 37-65). San Francisco: Jossey-Bass. Smith, E. M., Ford, J. K., & Kozlowski, S. W. ]. (1997). Building adaptive expertise: Implications for training design. In M. A. Quinones & A. Dudda (Eds.), Training for a rapidly changing workplace: Applications of psychological research (pp. 89-118). Washington, DC: American Psychological Association. VanLehn, K. (1996). Cognitive skill acquisition. Annual Review of Psychology, 47, 513-539. Welsh, E. T., Wanberg, C. R., Brown, K. G., & Simmering, M. J. (2003). E-learning: Emerging uses, empirical results and future directions. International journal of Training and Development, 7, 245-258. Wisher, R. A., 6k Curnow, C. K. (1999). Perceptions and effects of image transmissions during Internet-based training. The American journal of Distance Education, 13, 37-51.
DESIGNING DISTRIBUTED LEARNING SYSTEMS
3 APPLYING A SOCIAL CAPITAL PERSPECTIVE TO THE EVALUATION OF DISTANCE TRAINING KENNETH G. BROWN AND MARK E. VAN BUREN
A number of published reviews have addressed the relative effectiveness of different types of training delivery technology, including comparisons between distance training by means of the Internet (or corporate intranet) and more traditional face-to-face training. These reviews generally conclude that there are no significant differences in reactions and learning across delivery options, provided the training is designed similarly in other respects (Clark, 1994; Kosarzycki, Salas, DeRouin, & Fiore, 2003; Russell, 2001; Welsh, Wanberg, Brown, & Simmering, 2003). Despite this finding, limitations in the research suggest that it would be premature to conclude that it does not matter whether training in organizations is conducted at a distance or face to face. The research covered by these reviews typically has compared learner reactions and learning outcomes across various delivery technologies; for example, do learners who receive instruction by means of some form of technology indicate the same level of satisfaction and perform as well on knowledge tests as learners who receive instruction from a person in a faceto-face setting? A close examination of this research reveals two important 41
limitations to a full understanding of the effect of distance training (compared with face-to-face training) in work organizations. The first limitation is that this area of research has primarily focused on individual reactions and learning, thus ignoring a number of other potential outcomes.1 In particular, training may influence other individual-level outcomes, such as trainees' attitudes toward the organization and relationships with other employees. This possibility has been, at least in part, captured by the concept of affective learning outcomes (Kraiger, Ford, & Salas, 1993). However, affective learning outcomes as described in the existing literature focus on individual-level attitudes and motivation and do not typically consider individual constructs that relate to relationship-building or higher level constructs such as the network of social relationships in an organization. We believe that this is an important oversight and that group- and organizational-level outcomes may be affected by training and, in particular, the choice of delivery technology. The second limitation is that research in this area is often atheoretical. Many studies about delivery technologies have asked a somewhat simplistic question: Which technology is better? This research often examines differences in learning across different delivery technologies but does little to explain why differences occur, if they occur at all (Clark, 1994; Welsh et al., 2003). We believe that greater use of theory is needed to explain why delivery technology does (or does not) make a difference. As part of this effort, we strive in this chapter to provide a relatively balanced answer to the question of which technology is better by answering with: It depends on what outcome you are hoping to achieve. In this chapter, we argue that the criteria used to evaluate training, particularly when distance training is being compared with face-to-face training, should be expanded. More specifically, we use social capital theory to suggest that organizational-level outcomes may be influenced by training characteristics, including the choice of delivery medium. Using social capital as a starting point, we develop a conceptual framework for researchers interested in studying the effects of training at the organizational level of analysis. Most relevant to the purpose of this volume, the framework can guide both theoretical and practical research on the possible effects of shifting training from traditional face-to-face instruction to distance training. The propositions that follow from the framework are unique in the training literature in both their focus on relationships and social processes, which are often ignored in this area, and their emphasis on organizational-level outcomes.
'This criticism is not limited to research on distance training; training research generally focuses on a limited range of individual-level outcomes (Feldman, 1989; Kozlowski, Brown, Weisshein, CannonBowers, &. Salas, 2000).
BROWN AND VAN BL7REN
This chapter is organized as follows: First, we review social capital theory to establish the importance of social processes as they relate to learning and performance in organizations. Second, we explore connections between the social capital concept and existing training research and present propositions. Third, we review research on the use of technology for communications along with propositions about the effect of delivering training via distance on the development of social capital. Finally, we offer a discussion that raises a number of issues to consider in future research.
SOCIAL CAPITAL The fundamental premise of the concept of social capital is that the goodwill others feel toward us has value (Adler & Kwon, 2002). As with physical capital (i.e., manufacturing equipment) and human capital (i.e., skilled workers), social capital facilitates productive activity. Goodwill has value because it facilitates the flow of information and influence among people. Although social capital can be defined at the individual (e.g., Morrison, 2002), organizational (e.g., Adler 6k Kwon, 2002), and societal levels (e.g., Putnam, 1995), our focus is at the organizational level, where social capital has been demonstrated to provide competitive advantage for organizations (Florin, Lubatkin, & Schulze, 2003; Pennings & Lee, 1998; Tsai & Ghoshal, 1998). Before we provide a formal definition of social capital, it may be useful to illustrate its value. To do so, we offer the following hypothetical example. Because of an established relationship, a senior manager receives a call about an impending new product release from a friend who works for a competitor. This manager immediately calls other managers in the company and because of the quality of their relationships, easily convinces them to cancel their weekend plans and meet to discuss how to react to this news. A sense of solidarity among the managers expedites changes to their organization's product mix and advertising strategy in response to the competitor's product release. In this example, the organization's response to a change in its competitive environment is expedient because of the goodwill that exists between one of its managers and a person external to the organization, and among the managers as a group. In short, the organization benefits from social relationships that exist within the organization and between its employees and people outside the organization (for additional examples, see Adler 6k Kwon, 2002; Coleman, 1988). As this example demonstrates, social capital is actually an umbrella concept that integrates benefits of positive social relations discussed in other areas of research, such as social network structure (Burt, 1992), trust (Tyler 6k Kramer, 1996), organizational citizenship A SOCIAL CAPITAL PERSPECTIVE
behavior (Podsakoff, MacKenzie, Paine, & Bachrach, 2000), job embeddedness (Mitchell & Lee, 2001), and cohesion (Beal, Cohen, Burke, & McLendon, 2003). The expansiveness of the concept explains much of its appeal, but it is also the source of confusion. For the purpose of developing propositions about the effects of training and technology on social capital, we define social capital in more detail in the next section. Forms and Dimensions Recent treatments of social capital in the management literature have distinguished forms (internal and external; Adler & Kwon, 2002) and dimensions of social capital (structural, relational, and cognitive; Nahapiet & Ghoshal, 1998). These distinctions are reviewed next. Forms Adler and Kwon (2002) categorized most definitions of social capital into two broad forms: (a) external and (b) internal. These two definitions and approaches are reviewed briefly next. In addition, benefits and drawbacks of each form of social capital are reviewed. External social capital focuses on social ties that bridge group members with other people external to their immediate social group. Definitions from this perspective include Portes's (1998)—"the ability of actors to secure benefits by virtue of membership in social networks or other social structure" (p. 6)—and Burt's (1992): "friends, colleagues, and more general contacts through whom you receive opportunities to use your financial and human capital" (p. 119). In these cases, social capital refers to information and influence that accrue to people and organizations that have social connections that others do not have. For example, a university whose staff have social ties to staff at an accrediting agency may have an easier time with accreditation than a university that does not have such ties. External social capital has benefits and drawbacks. The benefits, as previously suggested, include access to information and resources that would otherwise be unavailable. By extension, individuals and groups with these ties gain power and influence relative to other individuals and groups who do not have these ties. A commonly cited drawback of external social capital is the time that it takes to establish and maintain social relationships (Adler & Kwon, 2002). Relationships require maintenance in the form of sustained contact, which takes away time from other tasks and activities (Leana & Van Buren, 1999). Internal social capital focuses on social ties that bond group members to each other within a social group. Definitions from this perspective include Fukuyama's (1995)—"the ability of people to work together for common
BROWN AND VAN BUREN
purpose in groups and organizations" (p. 10)—and Putnam (1995): "features of social organization such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit" (p. 67). In these cases, social capital refers to the solidarity or cohesion that is felt among people within a closely connected group. For example, a university with senior officials in each of its colleges who trust one another may be better able to cooperate in a time of fiscal crisis compared with a university in which such trusting relationships do not exist. Internal social capital also has benefits and drawbacks. The benefits include information exchange within the organization, cohesion that facilitates compliance with group norms, and collaboration toward accomplishing group goals. In recent research, these benefits are assumed to be the causal mechanisms underlying the positive effects of social capital (Fischer & Pollock, 2004; Tsai & Ghoshal, 1998). The primary drawback of internal social capital is insularity from high levels of cohesion, which may limit constructive conflict and innovation (Adler & Kwon, 2002; Leana & Van Buren, 1999). Dimensions
Nahapiet and Ghoshal (1998) described social capital along three dimensions: (a) structural, (b) relational, and (c) cognitive. The structural dimension refers to the relationships or network ties among actors. It is the most basic dimension, as relationships among people are a necessary precursor to the use of those relationships for information and/or influence. The relational dimension refers to the nature of the relationships, most notably the degree to which they are characterized by trust and norms for reciprocity. The cognitive dimension refers to shared paradigms that facilitate common understanding and collective action. Along with the relational dimension, the cognitive dimension of social capital addresses what Leana and Van Buren (1999) referred to as associability, which is the willingness of employees to subordinate individual goals for the good of the organization. This willingness is important because it helps ensure that an employee's relationships (the structural dimension) will be leveraged to the benefit of the organization (as opposed to solely for the benefit of the employee). From this perspective, the structural dimension of social capital describes relationships that could benefit the organization, and the relational and cognitive dimensions describe conditions under which employees are likely to use those relationships for the organization's benefit. In a recent empirical study, Tsai and Ghoshal (1998) examined the relationships among these three dimensions of social capital and their effects on resource exchange among business units within a large firm. They focused on social ties as a measure of the structural dimension, trust as a measure
A SOCIAL CAPITAL PERSPECTIVE
of the relational dimension, and shared vision as a measure of the cognitive dimension. Social ties between business units and trust predicted resource exchange, which was operationalized as the provision of valued information, service, and support. Shared vision influenced resource exchange by means of increases in trust. Thus, this study demonstrated that each dimension of social capital is useful to organizations because it facilitates resource exchange. Summary The concept of social capital argues that certain relationships have value and can benefit organizations because they facilitate the flow of information and influence. Social capital can be conceptualized at different levels and can be characterized in two distinct forms (internal and external) and as having three distinct dimensions (structural, relational, and cognitive). In the following section, we address the degree to which training can influence the dimensions of social capital in organizations. To keep the number of propositions manageable, we do not formally present propositions for external social capital, although many of the points that follow are applicable to both forms of social capital. This issue is addressed in the Discussion section.
TRAINING AND SOCIAL CAPITAL Training has historically had as its primary concern the development of human capital. Human capital refers to people and their ability to be economically productive (Coleman, 1988). The purpose of training, after all, is to develop employee knowledge and skills so that employees are better able to perform their jobs. Of course, training can have many different objectives, and any particular training program should be evaluated on the basis of those objectives (Goldstein & Ford, 2002; Noe, 2005). For example, some training programs have an objective to improve or otherwise alter employees' attitudes, and in those instances the trainees' attitudes should be evaluated. Even from this perspective, prior training literature does not typically address the degree to which ancillary and perhaps unanticipated outcomes may arise from training (Whiting & Dreher, 2006). In that regard, Coleman's (1988) classic work argues for the importance of considering the impact that any intervention may have on social processes within organizations. What is currently known about the effect of training on the social processes subsumed under the concept of social capital? The answer is, unfortunately, very little. Discussions of the concept of social capital in the
BROWN AND VAN BUREN
context of human resources practices have ignored the role that training might play (Leana & Van Buren, 1999; Marsden, 2001). The key point we raise in this section is that training can affect the structural, relational, and cognitive dimensions of social capital. The measures Tsai and Ghoshal (1998) used—social ties, trust, and shared vision—are used here, along with reciprocity norms, to develop testable propositions about which training characteristics are most likely to influence these dimensions of social capital. Training characteristics that are considered in this section include the participants (i.e., who is trained?) and the training methods (i.e., what is trained, and how?). We discuss the influence of training technology (in particular, distance training by computer or face-to-face instruction) in the next section. Because of space limitations in this chapter, the propositions we offer are intended to be illustrative rather than exhaustive. The Structural Dimension: Creating Social Ties Training offers one means to create relationships that would not exist otherwise. In short, training provides an opportunity for people to meet and interact. Such interaction need not be limited to members of immediate work groups; organizations can invite employees across functional areas of the organization and from various levels of management.2 Thus, organizations can use training to foster network ties and create a denser social network. The first training characteristic that determines the extent to which training will influence network density is overall breadth of participation. Organizations vary in the degree to which they offer training to all employees or select groups of employees, such as front-line employees or managers (Sugrue, 2003). In general, organizations that bring a greater percentage of their workforce through training programs should offer more opportunities for relationship development than organizations that train only a few select employees. PROPOSITION 1: Organizations that involve larger percentages of their employees in training will have denser social networks than organizations that involve smaller percentages of their employees in training.
Of course, it is unlikely that all training is created equal with regard to fostering relationships. We believe that the timing of the training matters as well. More specifically, we argue that training that occurs during employee role transitions, when employees are most open to support, is most likely to create new relationships. Once established, habits drive information-
* Organizations can also invite suppliers, contractors, and customers to training, but this is related to the creation of external social capital, which we have set aside for now to keep the number of propositions manageable.
A SOCIAL CAPITAL PERSPECTIVE
seeking and decision making by individuals in work groups (Gersick & Hackman, 1990). When employees are moved into a new role and work group, they are likely to refrain from powerful habits, at least temporarily, that constrain their social interaction and prevent them from interacting with new people and developing new relationships. Once they settle into a position and work group, however, their interaction patterns once again become entrenched and less prone to change. As a result, employees are less open to relationship development when it is offered at other points in their careers. Prevalent training programs that occur at career transition stages in' elude orientation programs (offered when an employee starts work with a new organization) and introductory supervisory training and leadership development programs (offered with promotion into a new role). At these points in time, employees are undergoing changes in their work-related identities and social networks (see, e.g., Hill, 1992), and they should be more receptive to training that enables them to cultivate new relationships, particularly with peers going through similar transitions. PROPOSITION 2: Organizations that provide more training during role transitions (entry and promotions) will have denser social networks than organizations that provide less training during role transitions. Other training characteristics relevant to network density are the extent and duration of interaction among trainees. Many theories of relationship formation indicate the importance of repeated interaction for lasting relationships to develop (e.g., Berger & Calabrese, 1975; Kelley, 1979). Of course, training programs vary in the degree to which they encourage trainees to get to know one another, discuss issues, and interact frequently. Some training programs, for example, begin with ice-breakers that encourage social interaction and require trainees to work together. Programs that do not have these activities offer little opportunity for trainees to develop relationships. Similarly, some organizations primarily offer short training programs that do not afford much time for trainees to get to know one another. Other organizations offer longer training programs that allow time for trainees to establish relationships. Both the extent and the duration of interaction should influence the degree to which training programs build relationships among employees. PROPOSITION 3: Organizations with training programs that encourage interpersonal interaction among trainees will have denser social networks than organizations that do not encourage interaction among trainees. PROPOSITION 4: Organizations with longer training programs will have denser social networks than organizations with shorter training programs.
BROWN AND VAN BUREN
The Relational Dimension: Building Trust and Reciprocity Norms The presence of social ties does not ensure that those ties will be used for the benefit of the organization (Leana & Van Buren, 1999). The nature of the relationships formed influences whether the organization will benefit from them. Two elements of the relational dimension that help ensure that relationships will benefit the organization are institutional trust and reciprocity norms. As with social capital, trust has been defined in many ways, and it occurs simultaneously at different levels of analysis (Tyler & Kramer, 1996). Generally speaking, trust is defined as a "psychological state comprising the intention to accept vulnerability based on positive expectations of the intentions or behavior of another" (Rousseau, Sitkin, Burt, & Camerer, 1998, p. 395). Of interest for organizational social capital, and related positive outcomes of information exchange and collaboration, is trust in the organization and organizational members by nature of their membership in the organization. This type of trust is referred to as institutional trust or category-based trust (Kramer, 1999; Rousseau et al., 1998). It is important to note that we are not specifically concerned with the creation of trust among trainees (such as what might be accomplished with a team-building exercise) but with building trust toward the organization and its members. It is this latter form of trust that is related to internal social capital. What characteristics of training programs would facilitate institutional trust? Social identity theory suggests that people who view themselves as similar to others will be more likely to trust them (Brewer, 1981). Thus, organizations that use training to create a shared organizational identity will establish a shared characteristic on the basis of which employees may initially trust one another. Kramer, Brewer, and Hanna (1996) specifically noted that strong socialization toward a common identity will create more trust among employees. PROPOSITION 5: Organizations in which training emphasizes a shared organizational identity will have greater levels of institutional trust than organizations in which training does not emphasize such an identity.
Hodson (2004) recently argued that institutional trust is based on two factors: (a) a set of supportive employment practices and (b) coherent and competent management of production. Examining 204 separate organizational ethnographies, Hodson found support for the idea that these factors predicted organizational outcomes related to internal social capital, including organizational citizenship behaviors, coworker infighting, and employeemanagement conflict. In other words, organizations that use supportive employment practices and manage effectively generally have employees who commit higher levels of discretionary effort to the organization and come A SOCIAL CAPITAL PERSPECTIVE
into conflict with each other and the organization less often. Training characteristics are relevant to Hodson's work in two ways: (a) providing training is one way organizations can support employees, and (b) offering effectively run training programs helps demonstrate and build managerial competence. First, organizations that provide greater amounts of training are likely to be viewed as supportive of their employees, and thus employees will be more likely to trust the organization. This hypothesis is reinforced by theory and research on perceived organizational support, which suggests that organizations that are perceived to support their employees (e.g., by providing opportunities to learn) create a sense of obligation in their employees to reciprocate and help the organization succeed (Eisenberger, Armeli, Rexwinkel, Lynch, & Rhoades, 2001). PROPOSITION 6: Organizations that provide more opportunities for training will have greater levels of institutional trust than organizations that provide fewer opportunities. The second way in which Hodson's (2004) work is relevant concerns managerial competence. All basic models of trust contain competence as a key factor (e.g., Rousseau et al., 1998), so organizations with competent managers should be more likely to be judged as trustworthy. Training is relevant here as both a symbol of managerial competence and a means to achieve it. First, the effectiveness of the design and administration of training may be interpreted by employees as an indicator of managerial competence. Training that is poorly designed, and poorly administered, may lead employees to question the competence of management in general. For example, the following scenario would likely damage employees' beliefs about managerial competence: A large training initiative frustrates employees because of administrative hassles to sign up and attend a session and because the material presented is not relevant to the employees' jobs. Second, training is relevant as a means to managerial competence. The provision of managerial training can improve the competence with which managers handle their jobs. Thus, the provision of management training by an organization can directly influence managerial competence and, thus, trust by employees in the organization. PROPOSITION 7: Organizations in which training is competently administered, designed, and delivered will have greater levels of institutional trust than organizations in which training is not competently administered, designed, and delivered. PROPOSITION 8: Organizations that offer more and higher quality management training will have greater levels of institutional trust than organizations that offer less and lower quality management training.
BROWN AND VAN BUREN
Reciprocity norms are another important relational element of social capital. Norms are general expectations about what is and what is not appropriate behavior, and reciprocity norms specifically refers to the general expectation that people return assistance they have received from others (Gouldner, 1960). Although there is a general societal norm for reciprocity, organizations can and do vary in the degree to which such norms are present and enforced. That is, organizations may vary in the degree to which employees engage in generous reciprocity. A strong norm for reciprocity should foster trust among employees because employees know that good deeds will be repaid at a later date. Feldman (1989) argued that training programs are well suited to influence the development of organizational norms because they can provide explicit statements about behavioral expectations, stories about critical events to establish precedents about what is appropriate behavior, and punishment for behavior that deviates from the desired norm. It is easy to see that training programs can influence norms in part by articulating desired (and undesired) behaviors as well as soliciting (and punishing) those behaviors within the training environment. Thus, training programs that emphasize reciprocity with other employees, and embed it within the training, such as with small-group discussion activities in which employees share information and ideas, should help establish and reinforce reciprocity norms. PROPOSITION 9: Organizations in which training encourages employees to help one another will have stronger reciprocity norms than organizations in which training does not encourage helping behavior.
Research on reciprocity in organizations suggests that employees are more likely to be committed to and work for the organization when the organization is perceived to have made a commitment to the employee (Eisenberger et al., 2001). Thus, when organizations are perceived to be committed to employees and exerting effort to advance employees' interests, employees should be more likely to engage in discretionary behaviors associated with social capital. Related to training programs, training and development opportunities are often seen as an important way that organizations can help employees, by enhancing their human capital. Thus, put more broadly, organizations that provide more opportunities for training should be viewed as giving something of value to the employees, so employees should feel an obligation to return something of value. PROPOSITION 10: Organizations that provide more opportunities for training will have stronger reciprocity norms than organizations that provide fewer opportunities.
A SOCIAL CAPITAL PERSPECTIVE
The Cognitive Dimension: Establishing Shared Vision Shared vision requires common language and perspectives by employees and entails having similar beliefs about the purpose and future of the organization. Commonality in language, perspectives, and beliefs increases the ease with which employees can communicate with one another and helps ensure that communications do not result in confusion and poor coordination. Developing a shared vision may be especially critical when organizations experience a tremendous influx of new employees, such as when undergoing a merger or an acquisition, having a period of rapid growth, or entering a new territory. During such periods, training programs can have a direct and powerful influence on the degree of commonality. First, and perhaps most obviously, training can be used to teach all employees, new and old, about the organization. PROPOSITION 11: Organizations that provide training focused explicitly on their mission, vision, beliefs, and values will have employees who share a common vision to a greater extent than organizations in which such training is not provided.
Learning about the organization is one element of becoming socialized (e.g., Chao, O'Leary-Kelly, Wolf, Klein, & Gardner, 1994), and new employee orientation programs often serve this function. Thus, organizations that provide new employee orientation can create common ground by increasing the degree to which all employees use the same language and possess similar values and beliefs (Klein & Weaver, 2000). PROPOSITION 12: Organizations that provide extensive new employee orientations will have employees who share a common vision for the organizarion to a greater extenr than organizations rhat do not provide new employee orientation.
Not all orientation programs are created equal with regard to forging shared perspectives. One means that organizations can use to develop a widely held set of beliefs is through strong socialization tactics, or tactics used to transition outside members to full membership in the organization. Ashforth and Saks (1996) and Bauer, Morrison, and Callister (1998) have argued that many characteristics of socialization programs are sufficiently correlated that programs can be described along a continuum from individualized (where new employees are treated as individuals and activities are informal, random, and variable) to institutionalized (where new employees are processed as a collective and activities are formal, sequential, and fixed). The use of institutionalized socialization practices obviously should lead employees to identify with their organization and share a common view of it.
BROWN AND VAN BUREN
PROPOSITION 13: Organizations in which the training of new employees is institutionalized will have employees who share a common vision for the organization to a greater extent than organizations in which the training of new employees is individualized.
Summary We have argued not only that training programs can influence social capital but also that the participants who attend training programs, and the methods trainers use in training programs, will determine the degree to which training influences dimensions of internal social capital. We turn next to research that explains how the use of technology for distance training may affect the development of social capital.
TECHNOLOGY, DISTANCE TRAINING, AND SOCIAL CAPITAL Research from various disciplines and levels of analysis suggests that technology-mediated interaction used in distance training may not be as useful as traditional face-to-face training in building trust, norms, and shared perspectives (e.g., Olson & Olson, 2000). To translate this research into propositions regarding distance training, we focus on distance training programs that rely solely (or at least primarily) on computer-mediated communication. So, for purposes of this chapter, we define distance training as computer-delivered training that occurs without the physical presence of an instructor or other learners. The Structural Dimension: Creating Social lies Putnam (2000) discussed the high hopes that some people place in technology for forging social ties. He challenged these hopes by arguing that the increased use of network technology is unlikely to build social capital. Putnam (2000) argued, first, that there is social inequality in access to network technologies. Thus, only the relatively affluent will use the technology, and this will reinforce existing social networks instead of building new ones. Second, the Internet may turn out to be a passive form of entertainment (like the television) rather than an active form of communication (like the telephone). Putnam (2000) reviewed evidence that television watching is associated with declines in social and civic participation; it eats away at the time available for building and maintaining social relationships. At this stage, it is unclear whether people will use network technology in ways that produce the same results as television. It is also unclear whether
A SOCIAL CAPITAL PERSPECTIVE
Putnam's (2000) concerns are equally applicable to communications at work as they are to communications at home and play. However, it seems plausible that not all employees in a company will have ready access to and experience with technology, and thus trainees' social ties after training may reinforce existing communication patterns instead of forging new ones (Orlikowski, 1992). There is also evidence from communications research that relationships started via distance are slow to develop and hard to maintain. When communicating using computer technology, there is considerable uncertainty because of a relative lack of social and nonverbal cues and the potential for feedback delays (Parks & Floyd, 1996). Uncertainty regarding how to behave, how the other person will behave, or how to explain the other person's behavior are all theorized to prevent or at least slow down the development of relationships (Berger & Calabrese, 1975). The concerns voiced by Putnam (2000) and others must be balanced against the ease with which communication can occur with communications technology. Electronic mail, in particular, is becoming increasingly prevalent as a means to keep in touch and help maintain social ties. Moreover, Parks and Floyd (1996) found that relationships can develop from online interactions, and people frequently expanded the relationships to include communication by other means (e.g., telephone, postal mail, and face to face). McKenna, Green, and Gleason (2002) found that some people developed rich and long-lasting relationships with people whom they had met only through computer. In a distance training environment, trainees can meet a broad range of people, and they can use network technology to maintain these connections. They may also begin communicating by other means more suitable to relationship development. Taken as a whole, technology is a mixed blessing when it comes to relationship development: It reduces transactions costs, so it allows for more communication, but it reduces the quality of that communication such that relationship-building and maintenance is more difficult. The net effect with regard to distance training is likely to be that the actual number of social ties in organizations that make heavy use of technology-delivered training (vs. face-to-face training) will not be that different. Rather than propose the null, we do not offer a proposition regarding distance training and social ties. The Relational Dimension: Building Trust and Reciprocity Norms Because the information transmitted through computers is less rich than the information transmitted face to face, nonverbal messages that are useful for building trust are not transmitted (Rocco, 1998). To make matters
BROWN AND VAN BUREN
worse, the anonymity of some computer-mediated communication makes cheating, reneging, and extreme language—"flaming"—more common occurrences (Putnam, 2000). Thus, when interacting with others at a distance, behaviors that reduce trust are more likely to occur. Thus, Putnam (2000) concluded that "building trust and goodwill is not easy in cyberspace" (p. 176). Rocco (1998) found that teams communicating electronically did not build trust, whereas those working face to face did. Aubert and Kelsey (2003) reported that teams working remotely and working face to face experienced differences in changes in trust over time. Trust between members of local groups went up over time as they worked together, but trust between members of a remote team went down over time. So, in addition to initially lower trust for remote teams, they had more difficulty developing trust over time. These effects suggest that it is less likely that trainees within a single distance training program will develop trusting relationships compared with trainees taking a similar course face to face. However, these results do not speak directly to the issue of institutional trust. Although it is possible that these individual trusting relationships will, over time, aggregate to influence institutional trust, such emergent effects would take considerable time to unfold. From our perspective, it is unclear whether direct effects on institutional trust would occur. Of course, referring back to the issue of managerial competence, it does seem possible that poorly designed distance training would reduce employees' perceptions of managerial competence. In other words, the quality of distance training is likely to have an effect. Because this is simply a restatement of Proposition 7, no formal proposition is offered with regard to distance training. Can norms be fostered by means of online training the way Feldman (1989) argued they can with traditional, face-to-face training? It seems unlikely. Feldman asserted that training would offer an opportunity for norms to develop because desired behavior could be explicitly described and rewarded, and undesirable behavior could be punished. Training online offers fewer opportunities for the observation of the full range of behaviors that are relevant to reciprocity. In other words, in distance training it may be possible to articulate the importance of sharing and reciprocity, but it is harder to have it demonstrated positively or negatively. In distance training, there are generally fewer opportunities for exchange and discussion. Even when such exchanges are made possible by text exchanges or video, they do not have the same vividness and intensity as face-to-face exchanges (Daft & Lengel, 1984). Consequently, it seems likely that organizations would be more likely to develop reciprocity norms through training when they make heavy use of face-to-face rather than distance training.
A SOCIAL CAPITAL PERSPECTIVE
PROPOSITION 14: Compared with organizations that make heavier use of face-to-face training, organizations that make heavier use of distance training will have more difficulty establishing and maintaining reciprocity norms.
The Cognitive Dimension: Establishing Shared Vision Putnam (2000) argued that the Internet allows people to confine their communications to people with whom they share common perspectives, and thus it encourages single-stranded, or tightly focused communication. Because it is relatively easy to walk away from an electronic mail message or a Web posting without attending to the message, ideas that people disagree with or do not understand can simply be ignored. The end result is that "local heterogeneity may give way to more focused virtual homogeneity as communities coalesce across space" (Putnam, 2000, p. 178). Where this limitation of communication technology is most relevant is in the socialization of new employees. Orientation programs can help people understand the language, values, and goals of the company (Klein & Weaver, 1998; Wesson & Gogus, 2005); however, Wesson et al. (in press) found that when orientations were conducted via computer, employees were less effectively socialized to the organization's vision and values. In other words, by shifting to computer delivery, the company was less effective at establishing shared values and vision. PROPOSITION 15: Organizations that orient new employees by means of distance training will have fewer employees who share a common vision for the organization than organizations who orient new employees face to face.
Summary The results of this review suggest that the increasing use of technology to deliver training programs may reduce the degree to which training builds trust, reciprocity norms, and shared vision and thus may hinder the development of internal social capital. By extension, organizations that shift much of their training to technology delivery may, over time, reduce the internal social capital that would otherwise have been created by face-to-face interactions.
DISCUSSION Training has been primarily conceived as a means to improve human capital, or the capabilities of the workforce. The social capital perspective
BROWN AND VAN BUREN
suggests that human capital is insufficient for understanding how organizations gain competitive advantage. In addition to individual capability, the nature of the relationships both within the organization and between members of the organization and external stakeholders plays an important role. The importance of relationships for people and organizations, and the fact that training and technology influence them, is not entirely new. There is a rich, historic tradition regarding the importance of social relationships in organizations (Blau, 1964; Homans, 1950) and, more specifically, how technology influences those relationships (Trist & Bamforth, 1951). In addition, the concept of affective learning outcomes (Kraiger et al., 1993), research on team training (Salas & Cannon-Bowers, 2001), and theory and practice in organizational development (French & Bell, 1998), all acknowledge that training can change participant attitudes and relationships. Nevertheless, the net influence of training, at a distance or in the classroom, on relationships has gone essentially unexplored. For example, a recent theoretical treatment of the relationship between human resources practices and social capital addressed the role of selection and placement, but not the role of training and development (Leana & Van Buren, 1999). Similarly, glossaries and indexes of popular training and development textbooks (Blanchard & Thacker, 2004; Desimone, Werner, & Harris, 2002; Goldstein & Ford, 2002; Noe, 2005) reveal no reference to the term social capital or related social processes. Even recent treatments of the relationship of training to types of capital beyond human and financial capital often focus on intellectual capital rather than social capital (Ardichvili, 2002; Van Buren, 2002). As a consequence, very little is known about how training influences the development of social capital. The theory and propositions presented in this chapter offer suggestions for future research and practice. The remainder of this discussion covers the research and suggestions for practice, with as great a focus on remaining questions as on implications. Research Implications The model offered here suggests a new set of training evaluation criteria. In addition to reactions and learning, researchers should examine changes in attitudes and beliefs related to social capital, such as institutional trust, reciprocity norms, and so on. Moreover, the effects of training programs on organizational, rather than just individual, outcomes should be considered. The theory and propositions raise a few more specific questions that bear further research. First, by focusing solely on social capital, we have assumed that an organization's choices about training characteristics (including delivery technology) do not differentially affect learning and thus the development of
A SOCIAL CAPITAL PERSPECTIVE
human capital over time. Although research supports the conclusion that well'designed training can stimulate learning with any delivery technology (Clark, 1994), it is possible that organizations systematically vary in how well they use particular technologies. Some organizations, for example, may have excellent classroom trainers. If the same organization hires novices with little instructional design experience to build online training, then it is unlikely that learning outcomes will be the same. The opposite may also hold true. Consider an organization that has considerable competence in building online systems and employees with experience using them; such an organization may obtain better learning with distance training. So, an interesting avenue for future research is to examine trade-offs between human and social capital development depending on firm competency, or previous experience, with various training design and delivery technologies. Second, we recognize that there may be moderators in the relationship between training characteristics and social capital. For instance, high levels of employee turnover may make it difficult to develop social capital regardless of training efforts. Likewise, training may be insufficient when social capital is very low, or redundant when social capital is very high. Thus, organizational characteristics such as high turnover rate and very high (or low levels) of social capital may diminish the effects that training has on the dimensions of social capital. Third, we did not address external social capital. Propositions could be developed about how organizations can use training to develop external social capital. In particular, involving customers, suppliers, and other stakeholders in training may prove to be one useful way for training to build relationships that connect employees to other organizations. Practice Implications In regard to practice, the theory suggests that organizations consider more than costs, reactions, and learning when designing training, in particular when changing delivery technology. To the degree that training has historically provided useful opportunities to develop trust, norms, and shared vision, large-scale transitions to distance training may have subtle but farreaching effects on the organization. Careful consideration of the effect of training on social processes is warranted in any deliberation of face-to-face versus distance training. By demonstrating the scope and scale of such effects, particularly to the organization's financial performance, a more thorough business case could be constructed to guide investments in distance training. This chapter raises one immediate practical question: How often do organizations convert courses that may be useful in developing social capital from collaborative, face-to-face formats to technology delivery that involves no collaboration? We not only believe this occurs but also have specific
BROWN AND VAN BUREN
examples: one from the literature and one from personal experience. First, in the literature, organizations have reported shifting orientation and socialization programs to the Web (e.g., Wesson & Gogus, 2005). Early orientation affords the foundation for creating social ties and building trust and norms. Thus, shifting orientation and socialization efforts to the Internet has the potential to hinder the development of internal social capital. Second, in our experience, we have witnessed a course high in collaboration shifted to a distance version that required no peer interaction. The course was a popular one taken by employees around the world. However, it often had a waiting list, and the costs of flying people to a common location to go through the training and complete the group exercises were seen as too high. The course was redesigned as a stand-alone one with no opportunities to meet and collaborate (Brown &. Gerhardt, 2002). It was, in short, the desire to increase "throughput" and decrease costs that led the organization to make this shift. In the absence of any theory or data suggesting this practice is problematic, organizations are unlikely to understand the full consequences of such decisions.
CONCLUSION The historical focus of research comparing training outcomes across delivery media leaves unanswered questions about the effect of wholesale transitions of training from the classroom to distance delivery. Prior research does suggest that distance training can reduce administration costs and, as long it is well designed, will otherwise be "not significantly" different from face-to-face training (Welsh et al., 2003). However, cost and individuallevel learning do not fully capture the outcomes related to training practices. In this chapter, we have built a theoretical framework for studying the effect of training on social capital as one important and generally overlooked outcome. Drawing on prior theory and research, we suggest that social capital can be influenced by training practices, including the choice of distance versus classroom training. We hope that the model and propositions we have offered will not only inform research on distance training but also broaden the focus of training research and evaluation to social as well as human capital.
REFERENCES Adler, P. S., & Kwon, S. (2002). Social capital: Prospects for a new concept. Academy of Management Review, 27, 17-40
A SOCIAL CAPITAL PERSPECTIVE
Ardichvili, A. (2002). The role of human resource development in transitioning from technology-focused to people-centered knowledge management. In C. M. Sleezer, T. L. Wentling, &. R. L. Cude (Eds.), Human resource development and information technology (pp. 89-104). Norwell, MA: Kluwer Academic. Ashforth, B. E., & Saks, A. M. (1996). Socialization tactics: Longitudinal effects of newcomer adjustment. Academy of Management journal, 39, 149-178. Aubert, B. A., & Kelsey, B. A. (2003). Further understanding trust and performance in virtual teams. Small Group Research, 34, 575-618. Bauer, T. N., Morrison, E. W., & Callister, R. R. (1998). Organizational socialization: A review and directions for future research. Research in Personnel and Human Resource Management, 16, 149-214Beal, D. J., Cohen, R. R., Burke, M. J., &. McLendon, C. L. (2003). Cohesion and performance in groups: A meta-analytic clarification on construct relations. journal of Applied Psychology, 88, 989-1004Berger, C. R., & Calabrese, R. J. (1975). Some explorations in initial interaction and beyond: Toward a developmental theory of interpersonal communication. Human Communication Research, I, 99-112. Blanchard, N. P., & Thacker, ]. (2004). Effective training: Systems, strategies, and practices (2nd ed.). Upper Saddle River, NJ: Prentice Hall. Blau, P. M. (1964). Exchange and power in social life. New York: Wiley. Brewer, M. B. (1981). Ethnocentrism and its role in interpersonal trust. In M. B. Brewer & B. Collins (Eds.), Scientific inquiry and the social sciences (pp. 345-360). San Francisco: Jossey-Bass. Brown, K. G., &. Gerhardt, M. W. (2002). Formative evaluation: An integrated practice model and case study. Personnel Psychology, 55, 951-983. Burt, R. S. (1992). Structural holes,. Cambridge, MA: Harvard University Press. Chao, G. T., O'Leary-Kelly, A. M., Wolf, S., Klein, H. ]., & Gardner, P. D. (1994). Organizational socialization: Its content and consequences, journal of Applied Psychobgy, 79, 730-743. Clark, R. E. (1994). Media will never influence learning. Educational Technology Research and Development, 42, 21-29. Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94, 95-120. Daft, R. L., &. Lengel, R. H. (1984). Information richness: A new approach to managerial behavior and organizational design. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior (Vol. 6, pp. 191-233). Homewood, IL: JAI Press. Desimone, R. L., Werner, J. M., & Harris, D. M. (2002). Human resource development (3rd ed.). Fort Worth, TX: Harcourt College Publishers. Eisenberger, R., Armeli, S., Rexwinkel, B., Lynch, P. D., & Rhoades, L. (2001). Reciprocation of perceived organizational support. Journal of Applied Psychology, 86, 42-51.
BROWN AND VAN BUREN
Feldman, D. C. (1989). Socialization, resocialization, and training: Refraining the research agenda. In I. Goldstein (Ed.), Training and development in organizations (pp. 376-416). San Francisco: Jossey-Bass. Fischer, H. M, & Pollock, T. G. (2004). Effects of social capital and power on surviving transformational change: The case of initial public offerings. Academy of Management Journal, 47, 463-481. Florin, J., Lubatkin, M., & Schulze, W. (2003). A social capital model of highgrowth ventures. Academy of Management journal, 46, 374-384. French, W. L., & Bell, C. H. (1998). Organisation development: Behavioral science interventions for organization improvement (6th ed.). Upper Saddle River, NJ: Prentice Hall. Fukuyama, F. (1995). Trust: The social virtues and the creation of prosperity. New York: Free Press. Gersick, C. ]., &. Hackman, J. R. (1990). Habitual routines in task-performing groups. Organisational Behavior and Human Decision Processes, 47, 65-97. Goldstein, I., & Ford, K. (2002). Training in work organisations: Needs assessment, development, and evaluation (4th ed.). Belmont, CA: Wadsworth. Gouldner, A. W. (1960). The norm of reciprocity. American Sociological Review, 25, 165-167. Hill, L. A. (1992). Becoming a manager: Mastery of a new identity. New York: Penguin Books. Hodson, R. (2004). Organizational trustworthiness: Findings from the population of organizational ethnographies. Organisation Science, 15, 432-445. Homans, G. C. (1950). The human group. New York: Harcourt, Brace. Kelley, H. H. (1979). Personal relationships: Their structure and processes. New York: Wiley. Klein, H. J., &. Weaver, N. A. (2000). The effectiveness of an organizational-level orientation training program in the socialization of new hires. Personnel Psychology, 53, 47-66. Kosarzycki, M. P., Salas, E., DeRouin, R. E., & Fiore, S. M. (2003). Distance learning in organizations: A review and assessment of future needs. In D. L. Stone (Ed.), Advances in human performance and cognitive engineering research (Vol. 3, pp. 69-98). Oxford, England: Elsevier Science. Kozlowski, S. W. J., Brown, K. G., Weissbein, D. A., Cannon-Bowers, J., &. Salas, E. (2000). A multi-level perspective on training effectiveness: Enhancing horizontal and vertical transfer. In K. ]. Klein &. S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations (pp. 157-210). San Francisco: Jossey-Bass. Kraiger, K., Ford, J. K., & Salas, E. (1993). Application of cognitive, skill-based, and affective theories of learning outcomes to new methods of training evaluation. Journal of Applied Psychology, 78, 311-328. Kramer, R. M. (1999). Trust and distrust in organizations: Emerging perspectives, enduring questions. Annual Review of Psychology, 50, 569-598.
A SOCIAL CAPITAL PERSPECTIVE
Kramer, R. M., Brewer, M. B., & Hanna, B. (1996). Collective trust and collective action in organizations: The decision to trust as a social decision. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organisations: Frontiers of theory and research (pp. 357-389). Thousand Oaks, CA: Sage. Leana, C. R., & Van Buren, H. J., III. (1999). Organizational social capital and employment practices. Academy of Management Review, 24, 538-555. Marsden, P. V. (2001). Interpersonal ties, social capital, and employer staffing practices. In N. Lin, K. Cook, & R. S. Burt (Eds.), Social capital: Theory and research (pp. 105-126). New York: Walter de Gruyter. McKenna, K. Y. A., Green, A. S., & Gleason, M. E. J. (2002). Relationship formation on the Internet: What's the big attraction? journal of Social Issues, 58, 9-31. Mitchell, T. R., & Lee, T. W. (2001). The unfolding model of voluntary turnover and job embeddedness: Foundations for a comprehensive theory of attachment. In B. M. Staw &. R. I. Sutton (Eds.), Research in organizational behavior (Vol. 23, pp. 189-246). London: JAI Press. Morrison, E. W. (2002). Newcomers' relationships: The role of social network ties during socialization. Acaaeirry of Management journal, 45, 1149-1160. Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational advantage. Academy of Management Review, 23, 242-266. Noe, R. A. (2005). Employee training and development (3rd ed.). New York: McGraw Hill. Olson, G. M., & Olson, J. S. (2000). Distance matters. Human-Computer Interaction, 15, 139-178. Orlikowski, W. J. (1992). Learning from notes: Organizational issues in groupware implementation. Proceedings of the ACM conference on computer-supported cooperative work (pp. 362-369). New York: ACM Press. Parks, M. R., & Floyd, K. (1996). Making friends in cyberspace. Journai of computermediated Communication, 1(4). Retrieved December 11, 2006, from http:// jcmc.indiana.edu/voll/issue4/parks.html Pennings, J. M., &. Lee, K. (1998). Human capital, social, and firm dissolution. Acadenry of Management journal, 41, 425-441. Podsakoff, P. M., MacKenzie, S. B., Paine, J. B., & Bachrach, D. G. (2000). Organizational citizenship behaviors: A critical review of the theoretical and empirical literature and suggestions for future research, journal of Management, 26, 513-563. Portes, A. (1998). Social capital: Its origins and applications in modern sociology. Annual Review of Sociology, 24, 1-24. Putnam, R. D. (1995). Bowling alone: America's declining social capital, journal of Democracy, 6, 65-78. Putnam, R. D. (2000). Bowling alone: The collapse and revival of American community. New York: Simon &. Schuster.
BROWN AND VAN BUREN
Rocco, E. (1998). Trust breaks down in electronic contexts but can be repaired by some initial face-to-face contact. CHI 1998 conference proceedings (pp. 496502). New York: ACM Press. Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. (1998). Not so different after all: A cross-discipline view of trust. Academy of Management Review, 23, 393-404. Russell, T. L. (2001). The no significant difference phenomenon: A comparative research annotated bibliography on technology for distance education (5th ed.). Montgomery, AL: International Distance Education Certification Center. Salas, E., & Cannon-Bowers, J. A. (2001). The science of training: A decade of progress. Annual Review of Psychology, 52, 471-499. Sugrue, B. (2003). State of the industry: ASTD's annualreview of U.S. and international trends in workplace learning and performance. Alexandria, VA: American Society for Training 6k Development. Trist, E. L., & Bamforth, K. W. (1951). Some social and psychological consequences of the longwall method of coal-getting. Human Relations, 4, 3-38. Tsai, W., 6k Ghoshal, S. (1998). Social capital and value creation: The role of intrafirm networks. Academy of Management Journal, 41, 454-476. Tyler, T. R., 6k Kramer, R. M. (1996). Whither trust? In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research (pp. 1-15). Thousand Oaks, CA: Sage. Van Buren, M. E. (2002). Making knowledge count: Knowledge management systems and the human element. In C. M. Sleezer, T. L. Wentling, 6k R. L. Cude (Eds.), Human resource development and information technology (pp. 105-130). Norwell, MA: Kluwer. Welsh, E. T., Wanberg, C. R., Brown, K. G., 6k Simmering, M. ]. (2003). E-learning: Emerging uses, empirical results, and future directions. International Journal of Training and Development, 7, 245-258. Wesson, M. ]., 6k Gogus, C. I. (2005). Shaking hands with a computer: An examination of two methods of organizational newcomer orientation. Journal of Applied Psychology, 90, 1018-1026. Whiting, S. W., 6k Dreher, G. F. (2006). Employee education and development programs: A framework for understanding unintended consequences. Unpublished manuscript, Indiana University, Bloomington.
A SOCIAL CAPITAL PERSPECTIVE
4 A META-ANALYTIC INVESTIGATION OF LEARNER CONTROL: OLD FINDINGS AND NEW DIRECTIONS KURT KRAIGER AND EDDIE JERDEN
This chapter addresses the role of learner control in distributed learning environments (DLEs). DLEs are those that use some form of technology (e.g., computers or videodiscs) to deliver training to individuals who may be separated from the instructional source by space or time. Common formats include computer-based instruction, computer-aided instruction, multimedia learning environments, intranet- and Internet-based instruction, Web-based instruction, and e-learning. In particular, we focus on learner-centered formats in which the software functions as a learning portal, exposing the learner to choices in both what and how to learn. For example, account managers may choose to read or listen to descriptions of new products, new sales techniques, or problems encountered by peers. Many modern forms of Internet-based training (or e-learning) follow this format. In this chapter, we address issues related to choices made by learners in such a flexible learning environment. We provide an updated meta-analysis of the learner control literature, derive a new model of learner control processes and outcomes, and suggest new avenues for research and application in the form of propositions.
UNDERSTANDING LEARNER CONTROL Learner control refers to the extent to which a learner can affect his or her own learning experience through control over features in his or her learning environment, such as the path, pace, or contingencies of instruction (Friend & Cole, 1990). In DLE contexts, learner control is thus distinguished from program control, in which the instructional software controls most or all of the decisions about what and how the trainee interacts with the training content (Hannafin, 1984). Program features that may or may not be under the control of the learner include topic choice, sequencing of information, pacing of information, on-demand assessment (i.e., testing at any time on a particular topic), capacity to modify screen design and text density, and opportunities for program advisement (Sims & Hedberg, 1995). Thus, learner control is intended to provide both customization of the content and the ability to accommodate learners' preferences for the amount or type of instruction (e.g., Freitag & Sullivan, 1995). For example, suppose that an employee is promoted to a team leader position and must take on project management responsibilities. She identifies an online project management training course. Her employer agrees to pay the fee for the course, but she must pay another important cost: her own time. As she begins to work through the online course late at night from home, she realizes that she already knows much of the fundamentals of project management and would like to skip ahead to specific knowledge and skills she needs for her job. Does the program allow her to do this? Is she correct in her assessment that she understands the basics? One of the useful features of the course she chose was the availability of streaming videos that use professional actors to illustrate principles covered in text. However, on some nights the new supervisor is tired and would like to skip the videos. On other nights, she wants to go back and review videos seen earlier in the training. Does the program allow her to do this? Does the program track where she is in the training so that if she backtracks, she is able to return exactly to where she departed? Can she modify how information is presented to her, for example, choosing to either read text online or listen to a narrator as she prepares herself for the next day? What type of feedback is available to her, and can she choose when she gets that feedback? Will she understand how to modify her approach to training if the feedback is negative, and will the program offer suggestions to help her with that decision? As illustrated in this example, it is important to understand that there are many ways in which learners may affect the instructional context. For the sake of brevity, it is useful to organize these characteristics into four broad categories (see Milheim &. Martin, 1991; Tennyson, Park, 6k Christen-
KRAIGER AND JERDEN
sen, 1985): (a) pace control, (b) sequence control, (c) content control, and (d) advisory control. Pace control allows learners to select the pace of work, for example, skipping easier material and dwelling on more difficult content. Sequence control allows learners choice in how to navigate course topics, for example, skipping topics or completing topics out of order. Content control allows learners to choose topics or assessments. Finally, advisory control refers to program-generated advice that informs learners of their progress or suggests a course of action. It is important to note that learner control may be provided in either DLEs or traditional classroom environments; much of the original learner control research was conducted in classroom settings (see Steinberg, 1977, 1989). However, because classroom instruction is typically tailored to the average learner and DLEs seek to provide individualized instruction (DeRouin, Fritzsche, & Salas, 2004), for the present purposes we restrict our investigation to DLEs. The previous example also highlights many of the proposed benefits of high learner control: principally, the opportunity to customize training content and training methods and the anticipated gains in trainee motivation and learning that result from that customization (Kinzie & Sullivan, 1989; Lepper, 1985). Here, trainee motivation refers simply to the desire to engage the learning software. Other proposed benefits of learner control include learning how to learn (Merrill, 1975) and opportunities for serendipitous learning (Staninger, 1994). Outcomes of Learner Control As noted previously, there are a number of hypothesized outcomes of learner control of interest to educational institutions or work organizations. These include effects on trainee learning (e.g., Gray, 1987) and trainees' reactions to instruction (Milheim, 1989) as well as other motivational, attitudinal, and cognitive outcomes, such as intrinsic interest in the training content (Becker & Dwyer, 1994), intrinsic interest in the learning process (Reigeluth & Stein, 1983), development of effective learning tactics and strategies (e.g., Merrill, 1975), and a willingness to explore the trained topic or other topics in greater detail (Kinzie & Sullivan, 1989). However, as we later show, research evidence supporting the impact of learner control on any of these outcomes is mixed (Hannafin, 1984; Milheim & Martin, 1991; Niemiec, Sikorski, & Walberg, 1996; Steinberg, 1977, 1989). With respect to specific outcomes, a number of studies have found evidence for more learning in learner control environments (e.g., Avner, Moore, & Smith, 1980; Gray, 1987; Kinzie, Sullivan, & Berdel, 1988), whereas other studies have reported greater learning in program control conditions (e.g., S. S. Lee & Wong, 1989; MacGregor, 1988; Morrison,
LEARNER CONTROL: OLD FINDINGS AND NEW DIRECTIONS
Ross, & Baldwin, 1992; Steinberg, 1977). Still, other studies have reported no significant differences between learner control and program control conditions (Arnone & Grabowski, 1992; Gray, 1987; Kinzie & Sullivan, 1989). Several studies have found evidence that providing learner control has positive influences on trainees' attitudes. For example, Becker and Dwyer (1994) found that learners reported greater intrinsic motivation when working on a computer-aided instruction (CAI) program with learner control compared with students working on a paper-based task. Although a few studies have found negative effects (Gray, 1987) or no effects (Arnone & Grabowski, 1992), most have reported that providing learner control resulted in more positive affective reactions to the learning experience (e.g., Hintze, Mohr, & Wenzel, 1988; Milheim, 1989; Morrison et al, 1992). In general, most of these studies used between-group designs, so evidence of positive attitudes may be more properly interpreted as "participants receiving learner control responded more positively than participants receiving program control," rather than "participants preferred learner control to program control." In contrast, Hintze et al. (1988) used a within-subject design by exposing Danish dental students to complete learner-controlled, partial learnercontrolled, and computer-controlled instructional situations. The researchers found that most of these students preferred partial or complete learner control. Finally, a number of studies have examined the effects of learner control on time on task, or how long learners choose to engage in learning tasks. Various studies have found that learner control lessons take more time to complete (e.g., Dalton, 1990; MacGregor, 1988; Shyu & Brown, 1992), less time to complete (e.g., Murphy & Davidson, 1991; Tennyson & Buttrey, 1980), or about the same amount of time to complete (e.g., Kinzie & Sullivan, 1989; Lahey et al., 1973). On reflection, it is difficult to be certain what the intended effect should be. If the benefit of providing learner control is greater efficiency, then learners should be expected to spend less time on task than with high learner control. If the intended benefit is greater interest in the learning task or instructional material (Kinzie & Sullivan, 1989), then learners given control may spend more time on task because they choose to explore the material in detail. Meta-Analyses of Learner Control Studies Partly in response to the variability in study outcomes, several metaanalyses and reviews have been conducted on the effectiveness of learner control over the past decade. A meta-analysis provides a quantitative review of prior published and unpublished studies; each study outcome is treated
KRAIGER AND JERDEN
as a data point, and outcomes can be averaged over studies to determine an overall effect. An unpublished meta-analysis on the efficacy of learner control in CAI by Parsons (1992) found a decrease in achievement of 0.04 standard deviations for students using computer-based programs that provided more learner control than alternative programs. A subsequent meta-analysis by Niemiec et al. (1996) concluded that although the learner control construct is theoretically appealing, the overall effects of learner control in CAI seem "neither powerful nor consistent." Although Niemiec et al.'s (1996) review is relatively recent, we elected to conduct an updated meta-analysis on learner control for four reasons. First, we anticipated that there may have been a substantial number of additional studies published since their review. In fact, although Niemiec et al. reported a total of 24 usable studies, we found an additional 11 published or unpublished studies dated 1996 or later. Second, Niemiec et al. included samples of all ages, and we wanted to focus exclusively on adult populations to be able to address learner control issues in work-related situations. Third, the nature of CAI is continually changing, and the impact of learner control on learning may vary depending on emerging training systems. In particular, hypertext and hypermedia systems popularized around 1990 (Ayersman, 1996; Borsook & Higginbotham-Wheat, 1992), and Web-based instruction introduced around 2000 (Rosenberg, 2001), offer interactive formats that seem to lend themselves better to exploration and customization by learners. Hypertext learning systems are ones in which users operate in an open and exploratory space in which information is structured using nodes and links (Conklin, 1987). Hypermedia systems include links that use multimodal methods of learning, including visual, auditory, and text-based information (Jonassen, 1989). Regarding the relationship of hypermedia systems to learner control, Marchionini (1988) wrote the following: "such a fluid environment requires learners to constantly make decisions and evaluate progress, thus forcing students to apply higher order thinking skills" (p. 9). We were curious as to whether learner control might be more effective given the growing popularity of flexible CAI systems (as well as learners' increasing familiarity with link-based navigation systems). Fourth, although Niemiec et al. investigated some moderators such as study source or sample characteristics, they did not investigate all moderators of interest to us; neither did they use state-of-the-art methods for evaluating the efficacy of moderators (Hedges & Olkin, 1985; Hunter & Schmidt, 2004). Accordingly, we conducted an updated meta-analysis on the effectiveness of learner control in computer-based training. Our primary research questions were how learner control affects learning outcomes and attitudes toward instruction. In addition, we were interested in the potential moderating effects of several research and training design variables as well.
LEARNER CONTROL: OLD FINDINGS AND NEW DIRECTIONS
META-ANALYSIS OF LEARNER CONTROL An attempt was made to locate, summarize, and analyze the results of all published studies, unpublished doctoral dissertations, and conference presentations reporting the effects of learner control on training outcomes using computer-based training. Only studies that included a direct comparison of CA1 with learner control to CAI without learner control were considered for inclusion in the analysis. Studies needed to use adult (18 years or older) participants to be included. The indexes used in the search were PsycINFO, ERIC, ABI/INFORM, and the Social Sciences Index for studies dated from 1989 through the present. The start date of 1989 was selected to coincide with the publication of Steinberg's (1989) review of the cognition and learner control literature. Key words used in the literature search were learner control and computer' based instruction, learner control and computer-assisted instruction, learner control and computer-based training, and learner control and computer-based [earning. Reference checks from popular review articles and previous learner control meta-analyses were conducted as well. Forty-nine published studies, 9 unpublished dissertations, and 9 conference papers were identified as usable through a review of abstracts. Of the 67 studies, only 30 were suitable for inclusion in the analysis.1 Of those excluded, 20 did not report information necessary to calculate an effect size, 2 were redundant with samples from other studies, and 15 did not directly compare the effects of learner control with the effects of no learner control. Data Coding The following variables were coded for each study: year published, group means and standard deviations, zero-order correlations for the relationships between group membership and training outcomes, sample size, sample demographics, and a variety of study quality variables (e.g., study design, threats to validity, sampling strategy). Moderator variables coded were source (published vs. nonpublished), type of training task (applied vs. educational), learner experience with the training task (none vs. at least some), computer experience (not reported vs. reported), sample (work vs. university), type of learning outcome (cognitive, skill based, or affective), type of learner control (pace, navigation, or instructional style), and length of training.
Studies included in the meta-analysis can be found in the references section of this chapter as asterisked entries.
KRAIGER AND JERDEN
Analytic Methods The studies included in this analysis contrasted two or more groups on learning and affective outcome measures. The effect sizes reported were a mixture of d- and r-type indexes. The most common d-type scenario was an evaluation of the mean difference between the group that received CAI with learner control and a group that received either no training or some alternative (e.g., program control). Studies reporting r-type effects provided findings in the form of a relationship between the dichotomous variable learner control versus program or alternative to learner control and a continuous outcome variable. For our analysis, d statistics were converted to pointbiserial correlations and then corrected for dichotomization prior to being included in the meta-analysis. In all cases, learner control was the treatment group. Thus a positive effect size reflects the positive effect of learner control on training outcomes. A total of 54 effect sizes were obtained from the 30 studies retained for the analysis. All outcome variables were coded in two ways. First, they were coded as either learning outcomes or affective outcomes. For the purpose of this analysis, affective outcomes refer to emotionally based reactions to the training (e.g., satisfaction, amount of anxiety during training). Second, measures of learning outcomes were categorized into four subcategories based on the goals of training and measure content. The four subcategories were (a) procedural knowledge, (b) declarative learning, (c) transfer, and (d) retention (see Kraiger, 2002). Procedural knowledge refers to learning information about "how," and declarative knowledge refers to learning information about "what." Transfer refers to a change in on-the-job behavior as intended by training. The final subcategory, retention, can be defined as the extent to which the application or frequency of the new behaviors learned during training is maintained over time. Huffcutt and Arthur's (1995) sample-adjusted meta-analytic deviance outlier analysis was computed on the unconnected effect size estimates. The criteria used for the cutoff point were a sample-adjusted meta-analytic deviance statistic at least 1.5 standard deviations above the mean and the value at which the first break in the scree plot analysis of the statistics occurred.2 Results are presented with both outliers removed and included for each analysis. We corrected for sampling error and error of measurement in the dependent variable at the individual study level using procedures described
When this cutoff was used, four studies were identified as outliers for the learning outcomes metaanalysis (Gist, Schwoerer, & Rosen, 1989; Jeffries, 2001; Pridemore & Klein, 1993; Simon &
LEARNER CONTROL: OLD FINDINGS AND NEW DIRECTIONS
by Hunter and Schmidt (2004). A sample weighted average correlation was computed and then corrected for unreliability in the dependent variable. Overall, reliability estimates were available for 22 of the 30 studies. For studies that did not report outcome reliabilities, the average of all reported reliabilities for that outcome variable was used. For the learning category, the average reliabilities were .70 for the overall analysis, .70 for declarative learning, .67 for procedural learning, .77 for transfer, and .69 for retention. For the affective category, the mean reliability was .91. Two indexes were used to assess the degree to which variation in the corrected correlations was due to statistical artifacts. The first is the 75% rule. According to this rule, if 75% or more of the variance is due to artifacts, then one can conclude that all of it is, on the grounds that the remaining 25% is likely to be due to artifacts not corrected for (Hunter & Schmidt, 1990). The percentage of variance accounted for by the artifacts corrected for—sampling error and error of measurement in the dependent variables in this case—is represented in the tables as Vart. The other test index was the credibility interval, also provided in the tables.3
RESULTS The results of the analyses both with and without outliers are shown in Table 4.1. On the basis of the size of the confidence intervals, the effect statistics show more stability without the outliers. With the outliers removed, the omnibus (or overall) results for the learning category go from a corrected r of .09 to. 11, with an interval supporting significance. The affective outcome results increase from .00 to .11. A one-way analysis of variance was used to determine whether effect sizes differed by the type of learning outcome. We conducted this procedure using all 54 observed effects and the type of learning outcome as the indepen' dent variable. To address variability in sample sizes and error variance, we weighted the effects by the inverse of the sampling error variance for the analysis (Hedges 6k Olkin, 1985). Overall, learning type did have a significant effect on the results of the training programs for both the full analysis, F(3, 41) = 4.64, p < .05, and the analysis with outliers removed, Werner, 1996), and three studies were identified for the affective outcomes meta-analysis (Gist et al, 1989; Maki, Maki, Patterson, & Whittaker, 2000; Simon & Werner, 1996). 'Whereas confidence intervals can be used to assess the accuracy of the sample size weighted mean effect size by constructing intervals around the standard error of the uncorrected mean effect size, credibility intervals should be used to determine whether moderators are operating using the corrected standard deviation around the mean corrected correlation to create intervals (Whitener, 1990). In general, if the credibility interval is sufficiently large or does include zero, then one can infer that the mean corrected effect size is probably the mean of several subpopulations determined by the influence of moderators (Whitener, 1990).
KRAIGER AND JERDEN
S • s CM I - CO CO CO OJ T- O T- r~ O O
co •* CM co in in CD Is- ^ CD in i- ^f CM CM o oo oo
1*1 1 & 0 0 &^ t 01 .2
f "1 1
•2 o ,, |
O) O C35 CD CD O T- i- CO O CM O
CO CO T- CO CM CM i- CO O CM
t •* ^j
CD in CD CO CO O
o o T- o -i- o
00 CD CM T-
T- O CM O i-
•g .0) Ta
D) to X
i ™ i? t3 g
CD CD CO CO "t CD T- T- CO CO O T-
O C CO
O) *^ *-
oo in co oo -i-
T- 1- •* O CM
o o0 t ° £
.E o o)3
g I-_J s?.® LO ^^ fll
CM Is- O i- •
CD CD 00 CO CO -SI-
CM Is- CM in CD
CM in CD m Is- ^ « a; ° £ Is- CO 0 CM Is- 5 0) en gj to m
g> > H -5
g 'co ™ 0 c ™ = ? 0 .9
S gl Q-tS
CD T- in T- t-
CD CO T-
CD CD -
-i— o> co T— r~m r^ i- CM CM i-
CO O O O ^ CO
ote. The 95% confidence interva imple size; r = sample weighted irrected correlation; V(vec) = weii tests were conducted on the ob!
CO CD CO i-m CO CM
CD CD 5
CM o •> Driver Long-term
Creating Fundamental Knowledge
Figure 11.3. Strategic goals for distributed teams and distributed team training.
transfer within teams (SG7) and the development of existing knowledge within skill-acquisition research focusing on teams (SG8). With respect to Fiore et al.'s approach to melding narrative theory with distributed team training, this could be pursued in such a way that differing knowledge outcomes are supported. Specifically, multiple trajectories would allow the narrative concept to be explored in a way that potentially produces new knowledge (SG9) or develops usable knowledge in support of team learning and development (SG10). Last, we have suggested how team cognition research, as presented in Cooke et al.'s chapter, can be supported with different trajectories in pursuit of short- and long-term research outcomes. Specifically, the strategic goals articulated for the theory put forth by Cooke et al. more broadly allow for the pursuit of knowledge associated with cognition and communication, and the differing trajectories can encourage both a deeper understanding of the issues they have uncovered (SG11) and the application of their findings (SG12). Considered in combination, the goals we have articulated and the trajectories we have illustrated help us conceptualize what can be a strategically based portfolio for distributed teams. In this way, we are able to move our understanding of distributed teams forward through simultaneous consideration of scientific and societal needs.
PROBLEMS AND POSSIBILITIES
Strategic Consideration of Cognitive Processes and Products in Distributed Learning Environments Chapter 8: Mayer Mayer effectively demonstrates the important empirical and theoretical strides made in our understanding of how multimedia implementations influence the learning process. On the basis of the nature of this research, we suggest that Mayer's work in multimedia and cognition falls within Octant 7 (short-term, need-driven creation of knowledge) of our strategic science space. Specifically, this has clearly been a need-driven approach in that Mayer has attempted to understand how multimedia implementations influence learning. Yet it is undoubtedly contributing to our fundamental understanding of this topic; that is, it is producing, or has produced, new knowledge related to learning. Finally, this work, although based on a long line of laboratory studies, is specific enough to support a shorter turnaround. As articulated in this volume, Mayer's ideas are ready for movement in differing directions of our research space, as the short-term outcomes that could be derived from his theory are apparent. Strategic Goal 13 (SGI3): Develop policy to transition multimedia learning theory to curiosity-driven research with a short-term perspective but aimed at creation of new knowledge. Example Tactical Objective: Support laboratory studies using multimedia learning theory in brain imaging research. Considering the potential strategic goals based on Mayer's work, in the case of SG13 we propose a trajectory that moves multimedia theory from Octant 7 to Octant 3. In particular, this theoretical approach can be used to investigate brain-based processing distinctions in the context of multimedia learning. Mayer's theorizing is supported by a broad set of behaviorally based studies. When this theory is coupled with recent methodological advances in neuroscience, we may be able to develop a fuller understanding of the neurophysiology associated with learning when multimedia implementations are involved. For example, recent work in understanding working memory and its underlying neurology can be considered in the context of the burgeoning discipline of neuroergonomics (e.g., Hancock & Szalma, 2003; Parasuraman, 2003). An overarching goal of neuroergonomics is to use our knowledge of brain function to better design learning and performance systems. In particular, "knowledge of how the brain processes visual, auditory and tactile information can provide important guidelines and constraints for theories of information presentation and task design" (Parasuraman, 2003, p. 6). Strategic Goal 14 (SGI4): Develop policy to transition multimedia learning theory to research with a short-term perspective that targets the development of existing knowledge.
FIORE AND SALAS
Example Tactical Objective: Support e-learning field research such that technologies based on multimedia learning theory are tested in complex industry-based training. With respect to SGI4, we suggest research that moves from Octant 7 to Octant 8. We have proposed Octant 8 as a target for this trajectory because the extant knowledge manifest in this theory has many components ready to be refined and shaped within specific areas so that it can become usable knowledge (e.g., e-learning training and/or technology specifications for a given domain). Chapter 9: Jonassen Jonassen eloquently argues in his chapter for the importance of understanding the nature and process of problem solving and lays out the theoretical issues that would support such research. On the basis of this theorizing, we place the ideas presented by Jonassen within Octant 5 (long-term, needdriven creation of knowledge) of our strategic science space. This theorizing contributes to fundamental gains in our understanding of problem solving, but it has a clear basis for need (i.e., better training of a competent problemsolving workforce). Within the context of understanding problem solving, we have trajectories leading the research community to short- and longterm time frames but with differing knowledge outcomes. Strategic Goal 15 (SGI5): Develop policy to support research on problem solving across a variety of task contexts through research with a longterm perspective while creating new knowledge. Example Tactical Objective: Support laboratory studies that examine problem-solving learning environments across well-structured and illstructured task contexts when interacting takes place in distributed environments. In considering the potential strategic goals with SG15, our trajectory moves us from Octant 5 to Octant 1. In this instance, research on problem-solving learning environments is encouraged across task contexts. Furthermore, this research is targeted for laboratory settings so that factors could be varied for the purposes of developing an understanding of how they alter problemsolving processes and address fundamental issues that are generalizable to distributed problem solving overall. Strategic Goal 16 (SGI6): Develop policy to investigate how technologies affect problem-solving effectiveness through research policy with a short-term perspective while creating usable knowledge. Example Tactical Objective: Support field research investigating how decision-support software can be used in problem-based learning environments in distributed engineering.
PROBLEMS AND POSSIBILITIES
With respect to SG16, the trajectory takes us from Octant 5 to Octant 8. With this trajectory, we have problem-based learning environments still as a research focus, but the aim is to explore how technologies can be developed to augment the learning and performance factors surrounding learning theory in these environments. The notion is that this is a clear setting through which to support a shorter turnaround and development of potentially useful knowledge. Chapter 10: Wisher and Graesser
On the basis of a strong foundation of research coming out of cognitive psychology, Wisher and Graesser present an important area of study on question-generation processes and how query-based systems are developable for more complex learning environments. Because their approach illustrates longer-term possibilities, they fall within Octant 5 (long-term, need-driven creation of knowledge). Specifically, their theory pursues fundamental understanding in the area of learning systems, and its translation requires added and varied forms of research. Nonetheless, they elaborate on a coherent set of research questions that, from the policy perspective, can be developed to help us understand the design and development of efficacious distributed learning environments. More specifically, question generation represents an important area of inquiry requiring both increases in fundamental understanding and a more immediate investigation of how extant knowledge can be developed and tested. As such, we have trajectories that can guide researchers along both short- and long-term time frames, each with differing knowledge outcomes. Strategic Goal 17 (SGI7): Develop policy to examine question generation through research with a short-term perspective while creating new knowledge. Example Tactical Objective: Support laboratory research investigating agent-based technologies that support social-cognitive processes during distributed learning.
When considering candidate strategic goals with SGI7, we propose a trajectory moving from Octant 5 to Octant 7. This would support research wherein question generation could be explored through more focused study, for example, on intelligent agents that support the type of guidelines and recommendations put forth by Wisher and Graesser. Specifically, SGI7 is aimed at creating fundamental knowledge in that it would enable an understanding of automated pedagogical agents capable of interacting within the questiongeneration processing rubric. Nonetheless, it would move us closer to implementation in that it could produce research outcomes more directly translatable in the shorter term. 258
FIORE AND SALAS
Strategic Goal 18 (SGI8): Develop policy to investigate how computational systems can support question generation through research with a long-term perspective while creating usable knowledge. Example Tactical Objective: Support information sciences research on intelligently managing question processes in distance learning. With respect to SG18, the trajectory moves us from Octant 5 to Octant 6. In this case, a broader line of study would be developed such that information sciences research could examine the numerous issues surrounding how the question-generation process would be managed and controlled within distributed learning. Summarising the Research Portfolio Possible for Cognitive Processes and Products
With respect to this section and what we illustrate in Figure 11.4, we see a strategic portfolio of research projects tailored to understanding a specific area of inquiry within a science of learning focusing on distributed environments. In particular, this portfolio is devised to explore an understanding of the cognitive processes and the cognitive products emerging within distributed learning environments. Thus, with the trajectories devised for Mayer's theory of multimedia implementations we have the potential
Developing Usable o , Knowledge °8
1 l^3'6 SG,4\
05 Creating Fundamental Knowledge
. n SO,/
Figure 11.4. Strategic goals for cognitive processes and products in distributed learning environments.
PROBLEMS AND POSSIBILITIES
for research to produce knowledge in a short- and long-term time frame. In this way, the practicing education and training communities are supported through the development of policy that examines differing outcomes targeting the creation of new knowledge (SG13) as well as the development of knowledge (SG14). With respect to the trajectories suggested for the theory put forth by Jonassen, there is an important distinction to consider. Specifically, the difference would be that SGI5 supports research that will contribute to creating fundamental understanding, whereas SGI6 pursues the development of extant knowledge in technologies relating to distributed engineering as a specific context. Finally, with respect to the theorizing of Wisher and Graesser, the user communities can benefit more rapidly through policy that supports short-term outcomes (SG17), but the science of learning in distributed learning and training would still examine more complex issues requiring a longer-term focus (SG18). By strategically conceptualizing a set of targeted goals and respective trajectories, we create a portfolio built on the basis of strong theorizing and that enables the solving of both scientific and societal goals.
SOME CONCLUDING THOUGHTS ON STRATEGIC THINKING IN SCIENCE AND TRANSLATING RESEARCH INTO PRACTICE We close this chapter with, again, a broader perspective, this time on the intimate relation between science and society and how research can help meet societal needs. It is important to note that many of the chapters in this volume tend to cluster within particular octants of our space. This is not surprising given that the nature of the funding that is supporting much of this research—that is, funding primarily provided by mission agencies—was targeting identified needs. Nonetheless, our goal with illustrating these within our research space was to illustrate how such funding not only helps us understand a given phenomenon but also provides innovative ideas that support the development of additional research trajectories throughout the strategic space. In particular, research beginning in a particular octant does not preclude movement in any direction out of that octant. To foster such movement, the scientific, educational, and industry communities should pursue a more integrated collaboration in the investigation of a science of learning in distributed environments. Within the context of our strategic research space we can begin to imagine how interaction between policymakers and stakeholders can provide input to this space and help guide resource allocations. From this broader perspective, Sarewitz (1997) described how democratic ideals need to be infused into the way researchers formulate their scientific endeavors and translate scientific findings to help meet societal goals. He called arguments that only scientists are qualified to
FIORE AND SALAS
identify scientific problems and societal priorities self-serving and politically unconvincing. Describing how activists have influenced biomedical research, he noted that "there is ample evidence that when scientists work cooperatively with knowledgeable activists from outside the research community, science as well as society can benefit" (p. 30). More specifically, we consider the suggestions of Guston (2000), who encouraged the pursuit of a reciprocal relationship whereby the communities that need their problems solved (e.g., education, industry) allow the scientific community to consider their problems in detail, while the scientific community is more willing to understand their problems and provide that help. Without the creation of legitimate mechanisms to allow public and/or user input into research and development priorities, this goal remains out of reach. To better serve society Guston argued for what he called collaborative assurance, a process requiring the input of organizations and institutions that enlist and/or include nonscientists. Guston suggested that the broader community of stakeholders needs to be better integrated into the federal research model. This is an important point, and the lack of stakeholder involvement may be an explanation for the poor translation of research findings in areas such as the science of learning. Specifically, the translation of research results in the psychological sciences in general, and the cognitive and learning sciences in particular, have been far weaker than the magnitude of data produced would suggest. Indeed, in an important article supporting a better application of the science of learning, Newcombe (2002) suggested that the ways we teach and learn need to be more informed by laboratory experimentation in human cognition just as the practice of medicine is informed by laboratory studies in biology. The fault lies not only with the user communities—that is, educators and trainers in need of a science of learning—but also with the learning science communities, perhaps for not adequately enlisting input from what Guston described as boundary organizations: "institutions that sit astride the boundary between politics and science and involve the participation of nonscientists as well as scientists in the creation of mutually beneficial outputs" (p. 34). The writings of Guston and Sarewitz (e.g., Guston, 2000; Sarewitz, 1997) go further and argue that even models based on Stokes's (1997) notion of use-inspired research and Holton and Sonnert's (1999) view of Jeffersonian science are still firmly entrenched in dated views of science policy (i.e., nondemocratic, patronage forms of thinking about funding). Although we rely on the thinking of Stokes and the likes of Holton and Sonnert, our approach is not necessarily at odds with notions of what Guston (2000) termed "collaborative assurance" or Sarewitz's (1997) democratic ideals for science policy. In particular, we do not espouse any form of patronage funding system. Furthermore, nonscientists from user communities or boundary organizations would play an important role in helping to define
PROBLEMS AND POSSIBILITIES
programs within, or move research toward, what we have conceptualized as need-based projects. In short, our aim was more to get the practicing scientific community as well as those in policy circles to think about science and society within the context of a science of learning for distributed environments and offer guidance as to how to work together to form scientific projects that are broad in scope. Thus, our strategic science space is merely one step in the direction of a conceptual structure for use by the scientific community and boundary organizations. Only by providing a structure that encourages translation of research will scientific gains be more probable. We have tried to illustrate how to redress problems of input and translation by suggesting different ways a science of learning in distributed learning can be conceptualized. An important additional point to recognize for a strategic approach to science is that a generation of scientists must be trained to both understand and embrace such a model. This is best explained in a recent report by Sabelli and Dede (2001) on how to integrate research and practice. Although discussed in the context of education research and practice, their points speak to a much broader range of science. In particular, they support a shift such that "the impact of research on education practice goes beyond 'transfer' and 'action research' toward reconceptualizing the relationship between scholarship and practice as instead a scholarship of practice" (p. 2). We fully agree with this, as it resonates with our strategic science space and attempts to link research programs to the user communities. However, Sabelli and Dede are not the only people arguing for a sturdier bridge between research and practice. Branscomb, Holton, and Sonnert (2002) similarly asked how it is that we can get a cumulative knowledge base on thinking and learning, and make it accessible to practitioners? We need new avenues for capturing the wisdom of practice, and we need a new kind of professional who can bridge the worlds of research and practice [italics added], (p. 409)
Perhaps by taking notions such as "collaborative assurance" in conjunction with funding models such as our strategic science space, and by ensuring that graduate education embraces such a culture of thinking, we can increase the probability that science and practice become indistinguishable. Finally, a necessary complement to developing graduate training that encourages this type of worldview is a fundamental shift in the way academic scientists themselves are rewarded. Just as funding sources must dictate general research areas and even encourage partnerships for disciplinary integration, university administrations must also become involved for real research translation to at least begin. This requires more than reliance on rewarding publications in oftentimes esoteric journals in which important ideas and findings are presented only to a small group of like-minded individ-
FIORE AND SALAS
uals. For example, university tenure review panels can begin to make research outreach (e.g., working with user communities and boundary organizations to facilitate translational components of one's findings) a serious part of the tenure-granting process. Similarly, professional organizations can become involved by providing outlets where fundamental research is considered in the light of particular societal needs. This includes continuation of publications such as the American Psychological Association's Journal of Experimental Psychology: Applied and the Association for Psychological Science's Psychology in the Public Interest. In short, what is required is for the academic, professional, and funding communities to nurture, encourage, and reward thinking in both understanding and use. Although this chapter was merely an armchair exercise in strategic thinking about a science of learning in distributed environments—that is, a type of what-if question using the contributions to this volume—our goal was really nothing more than this. Writing the concluding chapter to edited volumes provides one with a unique opportunity to question and expand on the variety of ideas presented. We hope we have stimulated the thinking of a number of important constituencies—current and future scientists, department chairs, policymakers, and even user communities—with the result being that some or all of these people are thinking more strategically about a science of learning in distributed environments. We all must recognize that modern scientists need to understand the intimate relation between science and society as well as how the knowledge they produce fits into the larger societal picture. This must be provided through training in graduate school and supported by the academic and funding communities. Within this context, as we stated at the beginning of this chapter, the scientific problems we are currently facing require multidisciplinary approaches. Within our strategic science space we have proposed trajectories devised in such a way that they require multidisciplinary teams if they are to meet the stated strategic goals; that is, they require coordinated scientific efforts cutting across disciplines. Only when such teams can be formed, and these issues are beginning to be addressed, can a science of learning begin the important process of solving the problems of, and realizing the possibilities for, the 21st century.
REFERENCES Branscomb, L. M. (1999, Fall). The false dichotomy: Scientific creativity and utility. Issues in Science and Technology, 16(1), 66-72. Branscomb, L. M., Holton, G., & Sonnert, G. (2002). Science and society. In A. H. Teich, S. D. Nelson, & S. J. Lita (Eds.), AAAS science and technology
PROBLEMS AND POSSIBILITIES
policy yearbook (pp. 397-433). Washington, DC: American Association for the Advancement of Science. Bransford, J. D., Brown, A. L, & Cocking, R. R. (2000). How people, learn: Brain, mind, experience, and school. Washington, DC: National Academy Press. Fiore, S. M., Rubinstein, ]., &. Jentsch, F. (2004). Considering science and security from a broader research perspective. International Journal of Cognitive Technology 9, 40-42. Guston, D. H. (2000, Summer). Retiring the social contract for science. Issues in Science and Technology, 17(1), 32-36. Hancock, P. A., & Szalma, J. L. (2003). The future of ergonomics. Theoretical Issues in Ergonomic Science, 44, 238-249. Hoffman, R. R., &. Deffenbacher, K. A. (1993). An analysis of the relations of basic and applied science. Ecological Psychology, 5, 315-352. Holton, G., & Sonnert, G. (1999). A vision of Jeffersonian Science. Issues in Science and Technology, 16(1), 61-65. Latham, G. P. (2001). The reciprocal effects of science on practice: Insights from the practice and science of goal setting. Canadian Psychology, 42, 1-11. Newcombe, N. S. (2002). Biology is to medicine as psychology is to education: True or false? In D. F. Halpern & M. D. Hakel (Eds.), Applying the science of learning to university teaching and beyond (pp. 9-18). San Francisco: Jossey-Bass. Parasuraman, R. (2003). Neuroergonomics: Research and practice. Theoretical Issues in Ergonomic Science, 44, 5-20. Rubinstein, J. (2002). Aviation security long-term theoretical human factors research. International Airport Review, 6, 49-54. Sabelli, N., & Dede, C. (2001). Integrating educational research and practice: Reconceptualizing goals and policies. How to make what works, work for us? Retrieved March 5, 2004, from http://www.virtual.gmu.edu/ss_research/cdpapers/policy. pdf Salzinger, K. (2003, Summer). Moving graveyards. Psychological Science Agenda, 16(3), 3. Sarewitz, D. (1997, Summer). Social change and science policy. Issues in Science and Technology, 13(4), 29-32. Stokes, D. E. (1997). Pasteur's quadrant: Basic science and technological innovation. Washington, DC: Brookings Institution Press. Thorndike, E. L., & Woodworth, R. S. (1901). The influence of improvement in one mental function upon the efficiency of other functions. Psychological Review, 9, 374-382. Vicente, K. J. (2000). Toward Jeffersonian research programmes in ergonomic science. Theoretical Issues in Ergonomics Science, I, 93-113.
FIORE AND SALAS
AUTHOR INDEX Numbers in italics refer to listings in the references.
Baldwin, W., 68, 88 Bamforth, K. W., 57, 63 Baraket, T., 99, 115 Bartlett, F. C., 122, 141 Bassok, M., 124, 142 Battiste, B., 101, 115 Bauer, T. N., 52, 60 Baxter-Magolda, M., 196, 205 Beal, D. J., 44, 60 Beccerra-Fernandez, L, 120, 142 Beck, I. L, 211,217, 228,230 Becker, D. A., 67, 68, 84 Belanich, J., 224, 228, 230 Belbin, R. M., 78, 84 Bell, B., 150, 166 Bell, B. S., 16, 17, 20, 22, 25, 26, 34, 35, 36, 37, 38 Bell, C. H., 57, 61 Berdel, R. L., 67, 87 Berg, G., 126, 141 Berger, C. R., 48, 54, 60 Biederman, L, 97, 114 Bill, R. L, 84 Bjork, R. A., 95, 115 Black-Hughes, C., 27, 36 Blanchard, N. P., 57, 60 Blau, P. M., 57, 60 Blodget, H., 5, 12 Bloom, B. S., 216, 217, 220, 230 Body, R., 122, 141 Boire, M., 178, 184 Boling, N. C., 84 Bond-Chapman, S., 122, 141 Bor, A., 27, 37 Borgers, S. B., 90 Borsook, T. K., 69, 84 Bouras, C., 26, 36 Bourne, L. E., 95, 115, 116 Bousfleld, W. A., 121, 141 Bove, W., 178, 183 Bower, G., 177, 184 Bower, G. H., 121, 122, 141 Bowers, C. A., 120, 141, 143, 152, 166 Brainerd, L. E., 216, 231
Abbott, H. P., 125, 141 Abelson, R. P., 122, 144 Abrahamson, C., 123, 141 Ackerman, P. L, 22, 37, 79, 83, 86, 98, 114
Acton, B., 120, 144 Adafe, V. U., 224, 230 Adler, P. S., 43, 44, 45, 59 Aleven, V., 226, 230 Allen, G. ]., 81, 83 Altman, M, 124, 145 American Educational Research Association, 221, 230 American Psychological Association, 221, 230 American Society for Training and Development, 5, 11 Anderson, J. R., 21, 35, 95, 98, 113, 114, 116, 124, 141 Anderson, L. W., 175, 182 Anderson, R. B., 23, 38, 172, 176, 178, 183 Andrews, D. H., 94, 114 Anthony, M. K., 103, 114, 115 Ardichvili, A., 57, 59 Armeli, S., 50, 60 Arnone, M. P., 68, 84 Arthur, W., 71, 86 Ashforth, B. E., 52, 60 Ashworth, A. R. S., Ill, 103, 114, 115 Atkinson, R. K., 97, 116 Aubert, B. A., 55, 60 Avanzino, S., 27, 37 Avner, A., 67, 84 Ayersman, D. ]., 69, 84 Azavedo, R., 223, 230
Bachrach, D. G., 44, 62 Baddeley, A. D., 95, 97, 114, 116 Baggett, W. B., 211, 231 Bailey, J. S., 97, 114 Bal.M., 122, 141 Baldwin, T. T., 95, 114
Branscomb, L. M, 262, 263 Bransford, J. D., 122, HI, 187, 205, 211, 216, 230, 239, 240, 264 Brayton, C., 11 Brewer, M. B., 49, 60, 62 Brookshear, J. G., 128, 141 Brown, A. L, 210, 211, 230, 232, 239, 264 Brown, J. S., 84, 79 Brown, K. G., 16, 38, 39, 41, 42, 59, 60, 61, 63, 82,84 Brown, S. W., 68, 86, 89 Bruner, J., 122, 130, 131, 132, 133, 137, 139, 140, 141 Bryman, A., 178, 183 Bunzo, M., 99, 116 Burke, C. S., 7, 12, 16, 39 Burke, M. ]., 44, 60 Burns, H., 79, 84 Burt, R. S., 43, 44, 49, 60, 63 Burwell, L. B., 84 Bush, V., 6, 11 Burtrey, T., 68, 90
Calabrese, R. J., 48, 54, 60 Callister, R. R., 52, 60 Camerer, C., 49, 63 Campbell, V. N., 78-79, 79, 84 Cannon-Bowers, J. A., 15, 22, 26, 36, 38, 39, 42, 57, 61, 63, 120, 131, 141, 143, 148, 162, 165, 166, 167 Capps, C., 79, 84 Cassell, ]., 125, 144 Caukwell, S., 152, 166 Champagne, M. V., 225, 233 Chandler, P., 33, 37, 87, 177, 183 Chao, G. T., 52, 60 Chapman, S., 124, 144, 210, 233 Chapman, S. B., 122, 142 Chase, W. G., 102, 115 Chen, M.-C, 113, 116 Cherney, R. J., 81, 83 Chi, M. T. H, 102, 115, 124, 142, 216, 230 Chiu, M. H., 124, 142, 216, 230 Cho, Y., 84 Choi, H., 152, 162, 167 Christensen, D. L., 66-67, 90 Churchill, D., 186, 205
Churchman, C. W., 203, 206 Ciardiello, A. V., 212, 217, 230 Clark, M. C., 121, 141 Clark, R. C., 29, 33, 36, 42, 182, 182 Clark, R. E., 41, 58, 60 Clawson, D. M., 95, 99, 115 Cocking, R. R., 211, 230, 239, 264 Code, S., 149, 167 Coelho, C. A., 122, 142 Cohen, D., 60 Cohen, R. R., 44, 60 Cole, C. L., 66, 85 Coleman, J. S., 43, 46, 60 Collins, A., 84, 79 Collins, J., 214, 231 Collis, B., 26, 36 Conklin, J., 69, 84 Contractor, N. S., 152, 166 Cooke, N. J., 148, 149, 150, 151, 152, 154-155, 163, 166 Cooper-Pye, E., 226, 231 Coorough, R. P., 85 Coovert, M. D., 23, 39 Cope, D., 120, 144 Corno, L., 81, 85 Coulson, R. L, 79, 89 Craig, S. D., 210, 212, 224, 227, 230 Craiger, J. P., 163, 167 Craik, F. I. M., 216, 230 Crismond, D., 191, 206 Cromley, J. G., 223, 230 Cronbach, L. J., 79, 85
Crowe, C. M., 207 Cuevas, H. M., 34, 36, 152, 166 Curnow, C. K., 27, 39, 229, 234
Daft, R. L., 55, 60 Dalton, D. W., 68, 85 Davidson, G. V., 68, 88 Dede, C., 262, 264 Deffenbacher, K. A., 239, 264 De Jong, T., 90 de Leeuw, N., 124, 142, 216, 230 Denning, S., 122, 130, 142 Department of Defense, 94, 115 Derek, K., 103, 114 DeRouin, R. E., 41,61, 67,85 Deshler, D. D., 85 DeShon, R. P., 26, 37 Desimone, R. L., 57, 60
Diaz, V. M, 85 Dibble, E., 192, 206 Dillon, T.J., 211, 223, 224, 231 Domino, G., 82, 85 Don, A., 126, 142 Donchin, E., 99, 115, 116 Donovan, S. S., 224, 233 Douglass, S. A., 113, 116 Downing, S. M., 221, 231 Dreher, G. R, 46, 56, 63 Druckman, D., 95, 115 Duffy, R. J., 122, 142 Duguid, P., 84, 79 Duncker, K., 101, 115 Dunlosky, ]., 223, 232 Durso, F. T., 153, 166 Dweck, C. S., 81, 85 Dwyer, C. A., 221, 231 Dwyer, D. J., 131, 142 Dwyer, M. M., 67, 68, 84
Edelson, D. C., 211, 231 Eggan, G., 99, 116 Egidi, G., 124, 143 Eisenberger, R., 50, 51, 60 Elliott, L. R., 23, 39 Ellis, J., 229, 234 Ellis, R. K., 5, 11 Emerson, M., 223, 231 Endsley, M. R., 152, 167 Entin, E. E., 163, 166 Eom, W., 81, 85 Erev, I., 97, 117 Ericsson, K. A., 95, 115 Evans, S., 11
Fall, R., 216, 233 Farr, M. J., 102, 115 Faux, T. L, 27, 36 Feldman, D. C., 42, 51, 55, 61 Feltovich, P. L, 79, 89, 102, 115 Finkelstein, N., 120, 142 Fiore, S. M., 6, 7, 9, 11, 12, 16,34,36, 39, 41,61, 120, 121, 140, 143, 148, 152, 162, 166, 167, 244, 264 Fischer, H. M., 45, 61 Fisher, J. B., 85 Fisher, S. L., 81, 85
Fitts, P.M., 98, 115 Flavell, J. H., 22, 36 Florin, J., 43, 61 Floyd, K., 54, 62 Foltz, P. W., 151, 166 Forcheri, P., 79, 85 Ford, J. K., 21, 22, 34, 36, 39, 42, 46, 57, 61, 81,85, 95, 114 Forrest-Pressley, D., 211, 233 Fowlkes, J. E., 131, 142, 143 Franks, ]. ]., 122, 141 Frederiksen, J. R., 99, 115 Freind, C. L., 66, 85 Freitag, E. T., 66, 82, 85 French, W. L., 57, 61 Frensch, P. A., 187, 207 Fritzsche, B. A., 67, 85 Fukuyama, F., 44, 61 Furugori, N., 214, 232
Gagne, R., 122, 143 Gallaway, M. A., 192, 206 Gallego, M., 125, 144 Gallini, ]. K., 176, 183 Gardner, P. D., 52, 60 Garner, R., 178, 183 Gerhardt, M. W., 59, 60 Gerrig, R. ]., 124, 143 Gershon, N., 124, 143 Gersick, C. ]., 48, 61 Gholson, B., 210, 224, 230 Ghoshal, S., 43, 44, 45, 47, 62, 63 Giat, L, 81, 83 Gick, M. L, 101, 102, 115 Gillingham, M., 178, 183 Gist, M. E., 71, 72, 85 Glaser, R., 21, 22, 36, 99, 102, 115, 116, 122, 124, 142, 143 Gleason, M. E. J., 54, 62 Goettl, B. P., 103, 114, 115 Goforth, D. J., 82, 86 Gogus, C. L, 59, 63 Goldman, J. A., 80, 87 Goldstein, I., 46, 57, 61 Good, T. L, 223, 231 Gopher, D., 97, 99, 101, 115, 117 Gordin, D. N., 211,231 Gordon, S. E., 216, 231 Gorman, J. C., 149, 151, 166 Gouldner, A. W., 51, 61
Govindasamy, T., 17, 18, 36 Grabowski, B. L., 68, 84 Graesser, A. C, 209, 211, 214, 215, 216, 217, 218, 220, 223, 224, 226, 227,230, 231, 232, 233 Gray, ]., 191, 206 Gray, S. H., 67, 68, 86 Green, A. S., 27, 38, 54, 62 Greer, ]., 214, 231 Grice, H. P., 179, 183 Gully, S. M., 34, 38 Guston, D. H., 261, 264 Guzley, R. M., 27, 37
Hacker, D. ]., 223, 232 Hackman, J. R., 48, 61 Haladyna, T. M., 221, 23] Hall, E. P., 192, 206 Hamid, A. A., 16, 33-34, 37 Hamilton, R. L., 211, 230 Hamm, H., 34, 38 Hammond, K. R., 189, 205 Hancock, P. A., 256, 264 Handoe, L, 122, J45 Hanna, B., 49, 62 Hannafm, M. ]., 66, 67, 78, 79, 82, 86 Harabagiu, S. M., 218, 232 Harel, K. H., 223, 23 J Harp, S. F., 178, 183 Harpin, P., 81, 89 Harris, D. M., 57, 60 Hart, S. G., 101, 115 Harter, D., 226, 227, 231 Hartley, L. L., 122, 143 Harward, H., 122, J41 Hassett, M. R., 86 Hatano, G., 22, 37 Healy, A. F., 95, 115, 116 Hedberg, J., 66, 89 Hedges, L. V., 69, 86 Hedlund, J., 151, 153, 166 Heiser, ]., 178, 183 Heller, M. A., 152, 166 Hemphill, L., 123, 143 Herman, D., 125, 126, 143 Hernandez-Serrano, J., 196, 205, 206 Herrnstein-Smith, B., 137, 143 Higginbotham-Wheat, N., 69, 84 Hill, L. A., 48, 61 Hinsz, V. B., 147-148, 166
Hintze, H., 68, 86 Hirumi, A., 125, 143, 144 Hodson, R., 49, 61 Hoffman, R. R., 239, 264 Hoffman, T. W., 207 Holbrook, J., 191, 206 Hollenbeck, J. R., 151, 166 Hollingshead, A. B., 27, 38 Holton, G., 239, 240, 242, 261, 262, 263, 264 Holyoak, K. J., 22, 37, 101, 102, 115, 116 Homans, G. C., 57, 61 Howell, A. W., 86 Hrymak, A. N., 207 Hsin-Yih, S., 86 Hu, X., 214, 231 Huff, M. T., 27, 31, 37 Huffcutt, A. L, 71, 86 Hughes, C. E., 124, 143, 145 Hughes, R. G., 97, 114 Hung, W., 189, 193, 206 Hunter, J. E., 69, 72, 86
Idzikowski, C., 95, 116 Ilgen, D. R., 151, 166 IMS Global Learning Consortium, Inc., 232 Ingaki, K., 22, 37
Jackson, T., 214, 231 Jacobson, M. J., 79, 89 Jeffries, P. R., 71, 86 Jehng, J. C., 202, 206 Jensen, P. J., 122, 143 Jentsch, F., 6, 11, 120, 142, 244, 264 Johnson, M. M., 31, 37 Johnson, S. D., 192, 205 Johnson, W. L., 227, 232 Johnston, J., 120, 142 Jonassen, D. H., 69, 86, 186, 187, 188, 189, 190, 191, 193, 196, 198, 200, 205, 206 Jones, W. E., 97, 114 Jordan, P., 226, 231
Kalyuga, S., 33, 37 Kanfer, R., 22, 37, 79,86, 98, 114
Kaplan, M., 189, 206 Keller, J. M., 80, 87 Kelley, H. H., 48, 61 Kelsey, B. A., 55, 60 Kerry, T., 224, 232 Kiekel, P. A., 150, 151, 163, 166 King, A., 124, 143, 210, 212, 219, 232 King, P. M., 203, 204, 206 Kintsch, W., 216, 217, 232 Kinzie, M. B., 67, 68, 87 Kiser, K., 15, 37 Kitchener, K. S., 203, 204, 206 Klein, H. J., 52, 56, 60, 61 Klein, J. D., 71,88 Kleinman, D. L, 151, 166 Knerr, B., 120, 144 Knowland, K., 125, 143 Koedinger, K. R., 226, 230 Kolodner, J. L, 191, 206 Konttinen, J., 124, 143 Kosarzycki, M. P., 7, 12, 16, 39, 41, 61 Kozlowski, S. W. J., 16, 17, 20, 21, 22, 25, 26, 34, 35, 36, 37, 38, 39, 42,61 Kraiger, K., 21, 22, 36, 38, 42, 57, 61, 71, 78, 80, 82, 87, 88 Kramer, R. M., 43, 49, 61, 62, 63 Kretschmer, M., 99, 116 Kucan, L, 211,230 Kufera, J. A., 122, 141 Kumar, V., 214, 231 Kwon, S., 43, 44, 45, 59
Lahey, G. F., 68, 87 Lajoie, S., 99, 116, 191, 206 Lampton, D., 120, 144 Langan-Fox, ]., 149, 167 Langfield-Smith, K., 149, 167 Langston, M. C, 211, 231 Larkin, J. H., 22, 38 Latham, G. P., 239, 264 LaVancher, C., 124, 142, 216, 230 Lea, M., 152, 167 Leana, C. R., 44, 45, 47, 49, 57, 62 Lee, J., 87 Lee, K., 43, 62 Lee, S. S., 67, 87 Lee, T. W., 44, 62 Lehman, D., 187, 206 Lehmann, F., 216, 232
Lempert, R., 187, 206 Lengel, R. H., 55, 60 Lepper, M. R., 67, 87 Lesgold, A. M., 99, 116, 121, 141, 191, 206 Lester, J. C., 180, 184, 227, 232 Levin, H. S., 122, 141 Levine, J. M., 152, 162, 167 Lewis, M. W., 124, 142 Liles, B. Z., 122, 142 Lintern, G., 103, 117 Lockhart, R. S., 216, 230 Longman, D. J. A., 97, 114 Lonn, S., 178, 183 Low, R., 181, 184 Lu, S., 226, 231 Lubatkin, M., 43, 61 Lynch, P. D., 50, 60
MacGregor, S. K., 67, 68, 87, 213, 232 Mackenzie, S. B., 44, 62 Maier, D. J., 87 Maiorano, S. ]., 218, 232 Maki, R. H., 72, 87 Maki, W. S., 72, 87 Mandinach, E. B., 81, 85 Mandler, G., 121, 144 Mandler, ]., 122, 144 Mane, A. M., 99, 116 Mantovani, G., 151, 153, 167 Marchionini, G., 69, 87 Mars, R, 177, 178, 183, 184 Marshall, R. R., 207 Martin, B. L., 66, 67, 88 Martin, D., 120, 144 Martin, M. J., 151, 166 Martinez, M. E., 221, 232 Marx, B., 27, 38 Masie Center, 5, 11 Mateas, M., 126, 144 Matejka, J., 122, 141 Mathias, A., 171, 183 Mathison, C., 125, 144 Mattoon, J. S., 87 Mautone, P. D., 179, 183, 184 Mayer, R. E., 8, 11, 23, 29, 33, 36, 38, 87, 171, 172, 174, 175, 176, 177, 178, 179, 180, 181, 182, 182, 183, 184, 187, 206 McCalla, G., 214, 231
McClelland, G. H., 189, 205 McCormick, E., 103, 115 McDaniel, R., 120, 121, 126, 139, 142,
143, 144 McGrath, J. E., 27, 38 McGrath, D., 87 McGuiness, Q, 193, 206 Mckenna, C., 125, 144 McKenna, K. Y. A., 27, 38, 54, 62 McKeown, M. G., 211,230 McLendon, C. L, 44, 60 McNatnara, D. S., 209, 211, 216, 231, 232 McPherson, J. A., 120, 144 Meagher, P., 214, 231 Meisel, S., 27, 38 Meister, C., 26, 39, 124, 144, 210, 233 Merrill, M. D., 67, 79, 88 Metcalf, D. S., 120, 121, 143, 144 Meyer, T. N., 99, 116 Micikevicius, P., 124, 145 Milanovich, D. M., 162, 167 Milheim, W. D., 66, 67, 68, 88 Miller, T. M., 99, 116, 117 Minsky, M, 126, 127, 144 Mitchell, T. R., 44, 62 Moe, M. T., 5, 12 Mohr, H., 68, 86 Molfino, M. T., 79, 85 Monty, R. A., 77, 88 Moore, C., 67, 84 Moreno, R., 33, 38, 178, 180, 184 Morrison, E. W., 43, 52, 60, 62 Morrison, G. R., 67-68, 88 Morrow, D. G., 122, 141 Moshell, J. M., 124, 145 Mousavi, S., 181, 184 Mullich, J., 6, 12 Mullins, M. E, 38 Mumpower, J., 189, 205 Murphy, M. A., 68, 88
Naas, C., 179, 184 Nahapiet, J., 44, 45, 62 Nason, E. R., 38 National Council of Teachers of English, 211,232 National Council on Measurement in Education, 221, 230 National Research Council, 211, 232
Newcombe, N. S., 261, 264 Newell, A., 186, 206 Newlin, M. H., 5, 12 Niemiec, R. P., 67, 69, 88 Nisbett, R. E., 187, 206 Noe, R. A., 19, 38 Norman, D., 122, 144 Novick, L. R., 102, 116
O'Donnell, A. M., 26, 38 Oehlert, M. E., 90 Ogata, H., 214, 232 Ohlsson, S., 21, 38 Olde, B. A., 226, 231 O'Leary-Kelly, A. M., 52, 60 Olkin, I., 69, 86 Olson, G. M., 53, 62 Olson, J. S., 53, 62 Ong, J., 79, 88 Orlikowski, W. ]., 54, 62 Orvis, K. L., 224, 230 Oser, R. L, 131, 142, 143
Paas, R, 29, 33, 34, 38, 181, 184 Page, W., 124, 143 Paine, J. B., 44, 62 Paivio, A., 177, 184 Palincsar, A. S., 210, 232 Parasuraman, R., 256, 264 Park, O. C., 66, 90 Park, S., 78, 88 Parks, M. R., 54, 62 Parsons, J. A., 69, 88 Parush, A., 34, 38 Pasca, M. A., 218, 232 Pattanaik, S. N., 124, 143 Patterson, M., 72, 87 Pea, R. D., 211, 231 Penner, L. A., 163, 167 Pennings, J. M., 43, 62 Perkins, M. R., 122, 141 Perlmuter, L. C., 77, 88 Person, N. K., 209, 214, 215, 217, 218, 220, 223, 227, 231 Phillips,]., 5, 12 Phillips, P. P., 5, 12 Philopoulos, A., 26, 36 Pintrich, P. R., 81, 88 Podsakoff, P. M., 44, 62
Pokorny, R. A., 192, 206 Pollock, T. G., 45, 61 Popper, K., 185, 206 Fortes, A., 44, 62 Post, T. A., 187, 190, 207 Postmes, T., 152, 167 Pounds, K., 125, 143, 144 Pressley, M., 211, 233 Pridemore, D. R., 71, 88 Priest, A. N., 5, 12 Prusak, L, 60 Puntambekar, S., 191, 206
Putnam, R. D., 43, 45, 53, 54, 55, 56, 62 Quarati, A., 79, 85
Raghavan, K., 99, J15 Rajaram, S., 95, 116 Rakow, E. A., 82, 89 Ramachandran, S., 79, 88 Rasmussen, J., 193, 206 Reese, F., 81, 89 Reeves, B., 179, 184 Reeves, T. C., 89 Regian, J. W., 99, J17 Reigeluth, C. M., 89, 67, 79 Reimann, P., 124, 142 Reiser, R. A., 81, 85 Renkl, A., 29,38, 97, 116, 181, 184 Rexwinkel, B., 50, 60 Rhoades, L., 50, 60 Rhodenizer, L, 120, 141 Rickard, T. C., 95, 99, 116 Rickel, J. W., 227, 232 Rivera, K., 152, 166 Robertson, M. M., 152, 167 Robinson, D. H., 84 Rocco, E., 54, 55, 63 Roediger, H. L., Ill, 95, 116 Rogoff, B., 226, 233 Roscoe, S. N., 95, 96, 113, 116 Rose, C, 226, 231 Rosen, B., 71, 85 Rosenberg, M. J., 69, 84 Rosenshine, B., 26, 39, 124, 144, 210, 212, 213, 219, 220, 232, 233 Ross, S. M, 68, 82, 88, 89 Rotter, ]. B., 81, 89 Rousseau, D. M., 49, 50, 63
Rubinstein, J., 6, 11, 244, 264 Rumelhart, D. E., 122, 144 Rumsfeld, D. H., 94, 116 Russell, T. L., 41, 63 Ryokai, K., 125, 144
Sabelli, N., 262, 264 Sabol, M. A., 229, 234 Saks, A. M., 52, 60 Salas, E., 7, 9, 12, 15, 16, 22, 23, 26, 34, 36,38,39,41,42,57,61,63, 67,85, 120, 131, 141, 142, 143, 148, 150, 152, 162, 165, 166, 167 Salzinger, K., 238, 264 Sandier, 1., 81, 89 Sarewitz, D., 260 Schank, R. C., 122, 123, 126, 127, 144, 211, 218,233 Schauble, L., 99, 115 Schiflett, S. G., 23, 30, 39 Schmidt, A. M., 22, 39 Schmidt, F. L., 69, 72, 86 Schnackenberg, H. L., 89 Schneider, W., 96, 97, 116, 117 Schreiber, D. A., 39 Schulze, W., 43, 61 Schumaker, J. B., 85 Schunn, C. D., 124, 141 Schvaneveldt, R. W., 154, 167 Schwartz, S., 189, 206 Schwoerer, C., 71, 85 Scielzo, S., 34, 36 Sebrechts, M. M., 223, 233 Seibold, D. R., 152, 166 Seidel, R. J., 229, 234 Sengers, P., 126, 144 Serfaty, D., 151, 163, 166 Shiffrar, M. M., 97, 114 Shope, S. M., 152, 163, 166 Shtub, A., 34, 38 Shute, V. ]., 99, 103, 115, 116 Shyu, H. Y., 68, 89 Siegel, D., 115 Sikorski, C., 67, 88 Silverstein, N. E., 89 Simmering, M. ]., 16, 39, 41, 63 Simon, A., 186, 206 Simon, D. P., 187, 206 Simon, H. A., 102, 115
Simon, S. ]., 71-72, 72, 89 Sims, R., 66, 89 Sims, V. K., 178, 184 Singley, M. K., 95, 116 Sistrunk, F., 97, 117 Sitkin, S. B., 49, 63 Slavin, R. E, 224, 233 Slavings, R. L, 223, 231 Smith, C., 26, 36, 67, 84 Smith, E. A., 38 Smith, E. M., 34, 39 Smith, M. U, 187, 206 Smith-Jentsch, K. A., 120, 144 Snow, C., 216, 233 Snow, R. E., 79, 85, 89 Snowden, D., 122, 130, 145 Sobko, K., 179, 184 Sohn, M.-H., 113, 116 Songer, N. B., 233 Sonnert, G., 239, 240, 242, 261, 262, 263, 264 Sparrow, S., 6, 12 Spears, R., 152, 167 Spencer, L, 81, 89 Sperry, L. L, 122, 145 Spires, H. A., 180, 184 Spiro, R. ]., 79, 89, 202, 206 Springer, L., 224, 233 Srinivas, K., 95, 116 Staninger, S. W., 67, 88 Stanne, M. E., 224, 233 Stapleton, C. B., 124, 145 Stasser, G., 120, 145 Stein, B. S., 187, 205 Stein, F. S., 67, 79, 89 Steinberg, E. R., 66-67, 68, 70, 89, 90 Steiner, I. D, 149, 167 Steinhoff, K., 177, 184 Sternberg, R. J., 187, 207 Steuck, K., 99, 116, 117 Stokes, D. E., 4, 6, 12, 239-240, 241, 242, 243, 261, 264 Stone, D. L, 7, 12, 16, 39 Stout, R. J., 148, 162, 163, 165, 166, 167 Sturm, B. W., 123, 145 Sueda, T., 214, 232 Sugrue, B., 47, 63 Sullins, J., 210, 230 Sullivan, H. J., 66, 67, 68, 78, 82, 85, 86, 87, 89 Summerall, S. W., 90
Sutterer, ]. R., 80, 87 Swaak, ]., 90 Swam, M. L., 223, 233 Sweller, J., 29, 33, 37, 38, 181, 182, 184 Szalma, J. L., 256, 264
Taatgen, N., 98, 117 Tannenbaum, S. L, 26, 36 Tapangco, L., 179 Tennyson, R. D., 66, 68, 90 Thacker, J., 57, 60 Thagard, P., 101, 102, 116 Thorndike, E. L, 95, 117, 251, 264 Timmons, C. W., 90 Timmons, P. L., 90 Togher, L., 122, 145 Toney, R. J., 21, 22, 34, 38 Toth, J., 214, 231 Trabasso, T., 122, 145 Trist, E. L., 57, 63 Troper, J. D., 216, 233
Tsai, W., 43, 45, 47, 63 Tsiatsos, T., 26, 36 Turner, M., 125, 145 Tutoring Research Group, 224, 227, 230 Tyler, T. R., 43, 49, 63
U.S. Department of Defense, 221, 233 U.S. General Accounting Office, 5, 12
Van Buren, H. J., Ill, 44, 45, 47, 49, 57, 62 Van Buren, M. E., 57, 63 Van der Meij, H., 215,233 Van Duyne, L., 120, 142 VanLehn, K., 21, 39, 209, 211, 226, 231, 233 Van Merrienboer, J. J. G., 34, 38 Vassileva, J., 214, 231 Vaucelle, C., 125, 144 Vaughan, S. L, 120, 145 Vegge, S., 178, 184 Ventura, M., 224, 230 Vicente, K. ]., 244, 264 Vidulich, M., 97, 98, 117 Volpe, C. E., 26, 36 Voss, J. F., 187, 190, 198, 207 Vygotsky, L. S., 226, 233
Walberg, H. J., 67, 88 Wanberg, C. R., 16, 39, 41, 63 Wang, A. Y., 5, 12 Washburn, D., 120, 144 Weaver, N. A., 52, 56, 60 Webb, N. M., 216, 233 Weil, M., 99, IJ5 Weissbein, D. A., 38, 42, 61 Welsh, E. T., 16, 17, 39, 41, 42, 59, 63 Wenzel, A., 68, 86 Werner, J. M., 57, 60, 71-72, 89 Wesson, M. ]., 56, 59, 63 Wetzell, K., 171, 183 Wheeler, J. L, 99, 117 White, B. Y., 99, 115 White, C, 178, 183 Whitener, E. M., 72, 90 Whiting, S. W., 46, 63 Whittaker, P. D., 72, 87 Whitten, S., 226, 231 Wightman, D. C, 97, 103, 117
Winne, P. H., 80, 81, 90 Winzenz, D., 121, 141 Wisher, R. A., 5, 12, 27, 39, 210, 224, 225, 229, 230, 233, 234 Wittenbaum, G. M., 120, 145 Wolf, S., 52, 60 Wong, S C. H., 67, 87 Wood, P. E., 207 Woods, D. R., 188, 207 Woodworth, R. S., 95, 117, 251, 264
Yano, Y., 214, 232 Yechiam, E., 97, 117 Yeh, Y., 97, 117 Yehene, V., 97, 117
Zeisig, R. L, 120, 144 Zimmerman, B. J., 81, 90 Zuniga, L, 5, 12
in DLT design process, 18 interactivity and, 26 preoccupation with, at expense of instructional design, 17-18 Basic research, 6-7, 239-244 Blended learning, student preferences and, 5-6
Academic settings, DLT utilization, 5 Active learning instructional supports, 34 question-asking and, 211, 212 Adaptive knowledge and skills characteristics, 22 in continuum of competencies, 20-21 team cognition, 163, 164 Advanced distributive learning constructivist foundation, 211 definition, 210 goals for questioning in, 210, 224, 229 learning objects in, 229 strategies for increasing question quality and frequency, 225-229 Advanced Distributive Learning initiative, 94, 229 Affective functioning in learner control outcomes, 82, 83 Affective learning outcomes, 42 Air Force Distributed Mission Training features, 94 rationale, 112-113 significance of, 94 Air traffic control, 97-98 Applied research, 6-7, 239-244 Argumentation, 203-204 Attentional processes cognitive load theory, 181 multimedia presentation, 178-179, 181 Attitude of learner, learner control and,
Causal explanation, multimedia presentation, 174 Chicken sexing, 97 Cognitive flexibility theory, 202-203 Cognitive processes and products, 10 cognitive disequilibrium, 226 cognitive load theory, 181 continuum of knowledge and skill competencies, 20-22 fidelity within DLT environments, 9, 93_94, 99-102, 113 information richness and, 33 phases of skill acquisition, 98 in question-formulation, 217 recommendations for strategic research, 256-260 shallow and deep knowledge, 216217, 221 social capital, 45 strong methods, 187 team cognition, 9, 148-151, 254 troubleshooting, 192-193 types and levels of cognitive representations, 217 uses of narrative, 122-123, 125-126 See also Information-experience richness; Learning processes Cognitive representations, 216-217 Communication computational structures for narrative, 126-127 computer-mediated, 151-152 cues for important information, 179 distributed team performance, 162, 165, 254
68, 82-83 Automatic learning, 98, 99 AutoTutor, 227 AWACS-AEDGE, 105-108, 112
Backward transfer of training, 94, 103104, 113-114 Bandwidth communication challenges, 27
Communication, continued human-computer interaction, 179-180 measuring team performance, 150-151 Communication richness, 27 technology and, 53-54 trust formation and, 55 Competencies, knowledge and skill developmental continuum, 20-21 DLT design considerations, 27-29 instructional design foci and, 21 physical fidelity of training in acquisition of, 98-99 shallow and deep knowledge, 216217,221 three-phase model of skill acquisition, 98 for troubleshooting, 192 Computational structures for narrative, 126-130, 139-140 Computer-assisted training conversational agents in, 227-228 learner control effects in, 68-74 See also Human-computer interface; Multimedia presentations Computer-mediated communication, 151-152, 165 Constructivism, 211 Contextual knowing, 197 Conversational agents, 227-228 Cost-effectiveness DLT, 5, 6 instructional considerations, 35 physical fidelity and, 95-96 preoccupation with, at expense of instructional design, 17 training efficiency, 95-98 Cues, 179
Debriefing narrative form rationale, 119-122 narrative systems for distributed simulation-based training, 130-139 Decision theory, 189 Declarative knowledge and skills characteristics, 21 in continuum of competencies, 20-21
DLT technologies, 23 learner control effects, 77 learning processes, 21 physical fidelity of training in acquisition of, 99 Deep knowledge, 216. See also Depth of knowledge Demand for DLT, 4 Department of Defense, 93 DLT policy, 94 Depth of knowledge, 216-217, 220, 223-224 Design of DLT application of theoretical model, 34-35 elements of theoretical framework for, 18-19 future research, 32-34 instructional supports, 33-34 integration of model elements in, 27-29 as multidisciplinary effort, 35, 237-238 need for theoretical framework, 16, 17, 18 neglect of instructional design issues in, 16 preoccupation with technical issues, 17-18 repurposing of existing instructional content for, 17 technology selection, 28-31 typology of instructional features, 19 Design problems, 190-191 Development, narrative skills, 123 Diachronicity of narrative, 133-134 Distributed Dynamic Decision-Making, 105-108, 112 Distributed learning and training (DLT), generally basic-applied research dichotomy, 6-7 current utilization, 4-5, 7, 15-16 definition, 4, 65 demand, 4 global trends, 3-4 psychology issues, 7-8, 11 significance of, as research topic, 238 DLT. See Distributed learning and training Duration of training programs, 48
Efficacy of DLT in absence of instructional design, 17-18 current understanding, 5 development of social capital, 5356, 58-59 learner control and, 67—69, 72—77 organizational outcomes, 42 outcome measures, 42 shortcomings of existing research, 41-42 theoretical framework for evaluating, 42 E-mail, 151-152 Encoding, 21 Essay questions, 221 Expert knowledge nature of, 21-22 recommendations for strategic research, 252 validation of training systems, 104, 112 Explanations, deep knowledge requirements for, 216 Explicit propositions, 217 Feedback, 121, 140 in question-asking training, 213, 226 Fidelity within DLT environments, 9 cognitive, 93-94, 99-102, 113 deep and surface structure, 101-102, 104, 108, 112 definition, 93 DLT technology features, 23-26 information-experience richness of technologies, 29-30 learning requirements, 26 phases of skill acquisition and, 98-99 physical, 93-94, 95-98, 102 recommendations for strategic research, 251-252 theoretical rationale, 95 training effectiveness and, 96-98 training efficiency and, 95-96 validation of training systems, 102103, 113 Geographic distribution effects, 147, 148-149, 151-165
Goals, instructional in advanced distributive learning, 211,229 DLT design considerations, 16, 18, 27-29 knowledge and skill competencies, 20-21 learning processes and, 22 outcome evaluation and, 46 problem solving skills, 186, 204 selection of instructional design features, 28 social learning, 31 in theoretical framework for DLT design, 18-23, 32 troubleshooting problems, 192 See also Strategic goals for DLT policy and research
Habits of behavior, 47-48 Human capital, 46. See also Social capital Human-computer interface, 126 computer voice, 180 question-answering facilities, 210, 211,214 social context, 179-180 Hypertext learning systems, 69
Immersion DLT technology features, 23-26 information-experience richness of technologies, 29-30 Independent knowing, 197 Industrial settings, DLT in, 4-5 Information-experience richness, 19 communication in DLT, 27 DLT design considerations, 27-29 DLT technology features, 23 recommendations for research, 246-247 research needs, 32-33 social learning and, 31 technology selection for, 29 Information processing, information richness and, 33 Information-retrieval systems, answer quality in, 214 Information richness theory, 246-247 Instructional supports, 33-34
Interactivity of instructional technology, 26 information-experience richness, 30-31 Interpersonal relationships communications technology and, 53-54 employee role transitions, 47-48 extent and duration of interactions in formation of, 48 use of narrative in, 123 See also Social capital
Learner control antecedents, 79-80 benefits, 67 categories of, 66-67 conceptual model, 77-79 definition, 66 learner characteristics and, 81-82 learner preferences, 78-80, 82-83 meta-analysis, 68-72, 83 moderator variables, 77-79 outcome factors, 80-83 outcomes, 67-68, 72-77 perceived control, 77-78 recommendations for research, 248-249 trainee attitude and, 68 Learning objects, 229 Learning processes building referential connections, 177-178 cognitive load theory, 181 cues for important information, 179 for declarative knowledge and skills, 21 DLT design considerations, 16, 18 instructional supports, 33-34 mental model construction, 175-177 phases of skill acquisition, 98-99 preprocess/postprocess, 120 procedural knowledge and skills, 21 progression of knowledge and skill competencies, 21, 22-23 question-generation learning, 212-213 role of questioning, 209, 211, 229 significance of, as research topic, 238 strategic knowledge and skills, 22
systems-policy analysis problem solving, 197-204 in theoretical framework for DLT design, 18-23, 32 troubleshooting problems, 193-197 uses of narrative form, 123-125 See also Problem-solving Literacy, use of narrative forms for learning, 125 Locus of control, learner control outcomes and, 81
Management, organizational trust and, 50 Memory feedback for learning, 121, 140 hierarchies, 140 Mental model construction, 175-177 depth of knowledge for, 217 in problem solving, 186-187 for troubleshooting, 192 Metacognition, 22, 212 instructional supports, 34 learner control outcomes, 81-82 metacommunication questions, 220 Metatagging, 186 Military training Airborne Warning and Control Systems task, 94, 104-112 appeal of DLT, 93 battlefield stress effects, 95 cognitive fidelity in, 99-101 Department of Defense DLT policy, 94 distributed simulation-based training, 130 distributed teams, 148 DLT utilization, 5 goals, 95 Mission training, 23-26 Multimedia presentations, 10 cognitive strategies, 182 conciseness, 173, 178-179 conserving aspects, 173, 181 cues for important information, 179 definition and characteristics, 173-174 DLT technology features, 23 explanative type, 174 features, 172 future research, 182
goals, 175, 181-182 information-experience richness of technologies, 29 neurophysiology, 256 rationale, 171 recommendations for strategic research, 256-257 referencing in, 173, 177-178 sociability aspects, 173, 179-180 systematicity, 172-173, 175-177
Narrative construction, 9 canonicity and breach in, 137-139 in cognition, 122-123, 125-126 computational structures for, 126130, 139-140 concept of inheritance in, 129, 131, 134, 136-137, 138 context sensitivity, 132-133 developmental aspects, 123 diachronicity, 133-134 distributed debriefing, 119-122 encapsulation in, 128, 131, 135 frames model, 127 future research, 140 intentional state entailment in, 134-137 in learning, 123-125 memory hierarchies and, 140 narrative systems for distributed simulation-based training, 130-139 negotiability, 132-133 polymorphism in, 128-129, 131, 134, 135-136 problem representation, 199-201 recommendations for strategic research, 252-253 referentiality, 131-132, 133 script model, 127 social interaction and, 123 story problems, 188 National Science Education Standards, 211 Neurophysiology, 256
Object-oriented programming, 127-129, 130, 253
Online instruction problem-solving instruction, 185186, 191, 204, 204. See also Problem-based learning environments story problems, 188 Organizational functioning density of social networks and, 48 as DLT outcome measure, 42 institutional trust, 49-50 learner control effects, 248-249 productivity, 248 reciprocity norms, 51, 55-56 recommendations for strategic research, 247-251 shared vision, 52-53, 56 social capital theory, 247-248 socialization practices, 52—53, 56 value of social capital, 43-44, 45, 57
Point & Query software, 223, 224, 227 Policy formulation, 10 basic and applied science research and, 239-244 strategic thinking in science and, 238-239 See also Strategic goals for DLT policy and research Problem-based learning environments, 186 design, 191-192 systems-policy analysis problems, 197-204 troubleshooting problems, 192—197 Problem representation, 198-201 Problem-solving, 10 attributes, 186-187 cognitive fidelity in training in, 101 context factors, 187, 198-199 decision-making problems, 189 design problems, 190-191 dilemma-type problems, 191 domain specificity, 187-188, 198-199 goals for instruction, 186, 204 problem characteristics, 187-191 problem complexity and, 187 problem dynamicity and, 187, 188 problem structure and, 187
Problem-solving, continued recommendations for strategic research, 257-258 shortcomings of online instruction, 185-186, 204-205 story problems, 188 strong methods, 187 systems analysis problems, 190 troubleshooting problems, 189-190 ubiquity of, 185 See also Problem-based learning environments Procedural knowledge and skills characteristics, 21 in continuum of competencies,
20-21 learner control effects, 77 learning processes, 21 phases of skill acquisition, 98 physical fidelity of training in acquisition of, 99 Productivity, organizational, 248 Program control, 66 Progression of knowledge and skill competencies, 20-21 Propositions, 217
Questioning, 10 answer quality, 214 asynchronous vs. synchronous learning environments, 213-214 cognitive disequilibrium and, 226 cognitive processes in, 217 collaborative question-generation, 224-225, 228 common-ground questions, 220 computer facilities for, 210, 211 in current DLT environments, 213-214 depth of knowledge and, 216-217, 220, 223-224 differences across settings, 209, 223 feedback, 213, 226 formats, 221, 222 generic question frames, 219-220, 227 goals for advanced distributive learning systems, 210, 216, 224 learner patterns, 223 metacommunication, 220
modeling, 227 multiple choice questions, 221-222 Point & Query software, 223, 224, 227 question-answering mechanisms, 210 question-generation learning, 212213, 214, 225 questioning the author, 228-229 question taxonomies, 217-221 recommendations for strategic research, 258-259 significance of, in learning, 209-210, 211, 229 sincere information-seeking questions, 210, 215-216, 220 strategies for increasing quality and frequency, 225-229 test questions, 221 training for, 212, 213
Role transitions, training during, 47-4
Schemas and scripts, 122, 127, 137-138 Self-efficacy, DLT efficacy and, 5 Self-perception, organizational trust and, 49 Self-regulated learning, 212 Shallow knowledge, 216. See also Depth of knowledge Sharable Content Object Reference Model, 229 SHERLOCK, 191 Simulations distributed debriefing, narrative forms for, 119-122, 130-139 DLT technology features, 23-26 Sincere information-seeking questions, 210 Social agency theory, 179-180 Social capital cognitive dimension, 45 creation of, in training, 46-53 dimensions of, 44, 45-46 DLT and development of, 53-56, 58-59 as DLT outcome measure, 42 external, 44 forms of, 44 institutional trust and, 49-50, 54-55
internal, 44-45 opportunities for research, 57—58 practice considerations, 58-59 reciprocity norms, 51, 55-56 recommendations for research, 247-
Synthetic task environments, 26 metacognition in, 34 See also Simulations Systems analysis problems, 190, 197204
248, 250 relational dimension, 45 shared vision, 52-53, 56 structural dimension, 45, 47-48 value of, 43-44, 46, 57 Social identity theory, 49 Social learning human-computer interaction, 179-180 in Web-based environments, 31 Space Fortress, 99-101 Spatial contiguity, 177 Spending, DLT, 5 Standards for performance and testing, 211 Standards for the English Language Arts, 211 Strategic goals for DLT policy and research cognitive processes and products, 256-260 fidelity effects, 251-252 future of psychology profession, 262-263 information richness theory, 246-247 learner control effects, 248-249 multimedia presentations, 256-257 narrative theory, 252-253 organizational learning, 251 problem solving, 257-258 purpose, 244-246, 263 question-generation process, 258-259 social capital theory, 247-248, 250 societal context, 260-262 tactical objectives and, 245 team process and performance, 251-255 trajectories concept, 245 Strategic knowledge and skills characteristics, 21-22 in continuum of competencies, 20-21 learning processes, 22 Student preferences, 5-6
Tactical science, 244 Team process and performance adaptive nature, 163, 164 collaborative question-generation, 224-225, 228 collective (shared mental) model, 149-150 computer-mediated communication, 151-152, 165 geographic distribution effects, 147, 148-149, 151-165, 253-254 interpositional knowledge, 164-165 measurement, 150-151, 153-1155 process-oriented holistic model, 150 strategic research, 251-255 team cognition, 148-151, 164-165, 254 unmanned aerial vehicle ground control, 147-148, 152-153 TeamTHINK, 224-225, 228 Team training, 9 advantages of lower communication richness in DLT, 27 challenges for DLT, 26-27 fidelity study using backward-transfer procedure, 104-112 interactive learning, 26 Technology, DLT applications, 23 computational structures for narrative, 126-130 current state, 4 as endpoint of DLT design process, 18 increased utilization and, 4, 16 information content, 23 information-experience richness, 29 instructional features of, 23 learner control, 65, 66 preoccupation with, at expense of instructional design, 17-18, 31-32 primary features, 23 range of, 23, 65
Technology, DLT, continued recommendations for strategic research, 251 selection, 19, 28-31, 32 social capital formation and, 53-56 in theoretical framework for DLT design, 19, 32 trust formation and, 54-55 unique features, 26 Temporal contiguity, 177-178 Test questions, 221-222 Theoretical basis for DLT design application, 33 essential elements, 18-19, 32 evaluation of training outcomes and, 42 implications for practice, 34-35 models, 8-9 need for, 16, 17, 18, 32, 35, 42 opportunities for research, 32-34 Timing of training, 47-48 Training, generally creation of social capital in, 46-53, 56-57 in development of institutional trust, 49-50
in development of shared vision, 52-53 duration, 48 Transitional knowledge, 197 Troubleshooting problems, 189-190, 192-197 Trust definition, 49 DLT and formation of, 54-55 institutional, 49 perceived competence and, 50 Unmanned aerial vehicle ground control, 147-148, 152-153 Utilization, DLT current, 4-5, 10-11, 15 reasons for increases in, 4, 15-16 Validation of training systems, 102-112 Verbalization, 98 Virtual reality, 30 Web-based environments, 31 future research, 182
ABOUT THE EDITORS
Stephen M. Fiore, PhD, holds a joint faculty appointment with the University of Central Florida's Cognitive Sciences Program in the Department of Philosophy and the Institute for Simulation and Training. He earned his PhD (2000) in cognitive psychology from the University of Pittsburgh, Learning Research and Development Center. He maintains a multidisciplinary research interest that incorporates aspects of cognitive, social, and organizational psychology in the investigation of individuals and teams, and he has published in the area of learning, memory, and problem solving. Dr. Fiore has received research funding from organizations such as the National Science Foundation, the Transportation Security Administration, the Office of Naval Research, and the Air Force Office of Scientific Research. Eduardo Salas, PhD, is professor of psychology at the University of Central Florida (UCF) where he was selected as a Trustee Chair Professor and holds an appointment as program director for the Department of Human Systems Integration Research at the Institute for Simulation and Training (1ST). Previously, he was the director of UCF's Applied Experimental and Human Factors PhD Program. Before joining 1ST, he was a senior research psychologist and head of the Training Technology Development Branch of the Naval Air Warfare Center Training Systems Division for 15 years. During this period, Dr. Salas served as a principal investigator for numerous research and development programs, including the Tactical Decision Making Under Stress program, which focused on teamwork, team training, decision making under stress, and performance assessment. Dr. Salas has coauthored over 300 journal articles and book chapters and has coedited 19 books. His expertise includes assisting organizations in how to foster teamwork, design
and implement team training strategies, facilitate training effectiveness, manage decision making under stress, and develop performance measurement tools. Dr. Salas is a fellow of the American Psychological Association, the Human Factors and Ergonomics Society, and a recipient of the Meritorious Civil Service Award from the Department of the Navy.
ABOUT THE EDITORS